You are on page 1of 139

Package ‘agricolae’

February 24, 2010
Type Package
Title Statistical Procedures for Agricultural Research
Version 1.0-9
Date 2010-2-24
Author Felipe de Mendiburu
Maintainer Felipe de Mendiburu <f.mendiburu@cgiar.org>
Suggests akima, klaR, MASS, cluster
Description Agricolae was presented on 28 August 2009 in the thesis “A statistical analysis tool for
agricultural research” to obtain the degree of Master on science, mention Systems Engineering,
of the facultad de ingenieria industrial y de sistemas - Universidad Nacional de Ingenieria,
Lima-Peru (UNI), being approved with the qualification of 18.14 in a scale from 0 to 20. The
thesis includes a satisfaction survey of the library, with an index quality of the software of 0.8 in
scale of 0-1. These functions are currently used by the International Potato Center (CIP), the
Universidad Nacional Agraria La Molina (UNALM-PERU), and the Instituto Nacional de
Innovacion Agraria (INIA-PERU). It comprises the functionality of statistical analysis into
experimental designs applied specially in field experiments in agriculture and plant breeding:
Lattice, factorial, complete and incomplete block, Latin Square, Graeco, Alpha designs, Cyclic,
augmented block, split and strip plot designs, comparison of multi-location trials, comparison
between treatments, resampling, simulation, biodiversity indexes and consensus cluster.
License GPL
URL http://tarwi.lamolina.edu.pe/~fmendiburu
Repository CRAN
Date/Publication 2010-02-24 15:03:04
1

R topics documented:

2

R topics documented:
agricolae-package
AMMI . . . . . .
AMMI.contour .
audpc . . . . . .
bar.err . . . . . .
bar.group . . . .
BIB.test . . . . .
carolina . . . . .
Chz2006 . . . . .
CIC . . . . . . .
clay . . . . . . .
ComasOxapampa
consensus . . . .
corn . . . . . . .
correl . . . . . .
correlation . . . .
cotton . . . . . .
cv.model . . . . .
cv.similarity . . .
DAU.test . . . .
DC . . . . . . . .
delete.na . . . . .
design.ab . . . .
design.alpha . . .
design.bib . . . .
design.crd . . . .
design.cyclic . .
design.dau . . . .
design.graeco . .
design.lattice . .
design.lsd . . . .
design.rcbd . . .
design.split . . .
design.strip . . .
disease . . . . . .
duncan.test . . .
durbin.test . . . .
fact.nk . . . . . .
friedman . . . . .
frijol . . . . . . .
genxenv . . . . .
Glycoalkaloids .
graph.freq . . . .
grass . . . . . . .
greenhouse . . .
grid3p . . . . . .

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

4
5
7
9
10
12
13
15
16
17
18
19
20
22
23
24
25
26
27
28
29
30
31
32
34
35
36
38
39
41
42
43
44
46
47
48
49
51
52
53
54
55
55
57
58
59

R topics documented:
growth . . . . . .
haynes . . . . . .
Hco2006 . . . . .
hcut . . . . . . .
heterosis . . . . .
hgroups . . . . .
HSD.test . . . . .
huasahuasi . . . .
index.bio . . . .
index.smith . . .
intervals.freq . .
join.freq . . . . .
kendall . . . . . .
kruskal . . . . . .
kurtosis . . . . .
lastC . . . . . . .
lineXtester . . . .
LSD.test . . . . .
LxT . . . . . . .
markers . . . . .
melon . . . . . .
montecarlo . . .
natives . . . . . .
nonadditivity . .
normal.freq . . .
ojiva.freq . . . .
order.group . . .
order.stat . . . . .
pamCIP . . . . .
paracsho . . . . .
path.analysis . . .
PBIB.test . . . .
plot.graph.freq . .
plots . . . . . . .
plrv . . . . . . .
polygon.freq . . .
potato . . . . . .
ralstonia . . . . .
random.ab . . . .
reg.homog . . . .
resampling.cv . .
resampling.model
rice . . . . . . .
RioChillon . . . .
scheffe.test . . .
similarity . . . .
simulation.model
sinRepAmmi . .

3
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

60
61
62
63
64
65
66
68
69
70
71
72
73
74
75
76
76
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
94
96
97
98
99
99
100
101
102
103
105
106
107
109
110
111

Scheffe. . . . . . . . . . . . . . . . . . . . . . . . . yacon . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . HSD. . . . .plot . . . . . . . . . . . . . . . . . . waerden. . . Analysis of genetic experiments: North Carolina designs. . . . . . . . . . . analysis Mother and baby trials (see data RioChillon). . . . . . . . . . . soil . . . . . . . . RCBD. . . . . . . . Bonferroni and other padjust. . . . . . . . . . . . . . . . . . . . . .freq sweetpotato . . .freq . . . . . . . . . . . . . . . . . . . . . . LinexTester. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . wilt . . Comparison between treatments: LSD. . Partially Balanced Incomplete Block. . . . . . . . . . . . . . . . . . . . . . . . . . . . .freq .test . . . . . . . . . Durbin. . . . Waller. . . . . . . . . . . . . . . . . . summary. . . . . . . . . . . . . . . . . Alpha design. . . . SNK. . CRD. . . . . . . . . . . . . . . augmented block. . . . . . . . . . . . . . . . . . . . . 112 112 114 115 117 118 119 121 122 123 124 125 126 127 128 129 131 133 134 136 Statistical Procedures for Agricultural Research Description This package contains functionality for the Statistical Analysis of experimental designs applied specially for field experiments in agriculture and plant breeding. . .graph. . . . . . . . Strip plot. . . . . . . . . . Resampling and simulation: resampling. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Duncan. . . . . . . Path Analysis. . Van Der Waerden. . . . . . . . . .model. . . . . . . . . . . . . . . . . . . . . .stat . .4 agricolae-package skewness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Student Newman Keuls SNK. . . . . . . . . . . . . . . . . . . . . . Comparison of multi-location trials: AMMI Stability (biplot and triplot). . . . . . . . . . . . . . . . . . sturges. . . . . . . . . . . . tapply. . . . . . . . . . . . . . . . . . . . . . . . wxyz . . . . . . . . . . . . . . . . . . . . . . Cyclic. . . . . . . . . . . . . . . Friedman. . . . . . . .0-9 2010-2-24 GPL Planning of field experiments: lattice. . . . . . . . . .nonpar . . . . . vark . . . Soil Uniformity: Smith’s . . . . . . . waller . . . . . . . . . . . . simulation. . . . . . . . . . . . . . . . . . . . . . . . .test . . . . . table. . . . . . stability. . . . . . . .model. . . . . . . . . . . . . . . . . . . . . . . stability. . . Latin Square. . . . . . waller. . montecarlo Ecology: Biodiversity Index. . .freq . . . . . . . . . . . . . . . . . . . . . Non parametric tests: Kruskal. . . . Index agricolae-package . . . . . BIB. . . . . . . . . . . .test . . . . . . Details Package: Type: Version: Date: License: agricolae Package 1. . . . . . . . . . . . . . . . . . strip.par . . . . . . . . . . Graeco. . . . split and strip plot Designs. . . . . . stat. . . . . . . . . . . . . . . . . . . Balanced Incomplete Block. . . . . . . . . . factorial. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

triplot graphs and analysis.simon@cgiar... Cluster Analysis: Consensus Cluster.mendiburu@cgiar.) Arguments ENV Environment GEN Genotype REP Replication Y Response MSE Mean Square Error number TRUE or FALSE graph "biplot" or "triplot" ..org> Team leader: Reinhard Simon <r. Y. Usage AMMI(ENV. MSE = 0. Descriptive statistics utilities: *. REP. this function generates biplot. .graph="biplot"...AMMI 5 Index.org> References International Potato Center (CIP) AMMI AMMI Analysis Description Additive Main Effects and Multiplicative Interaction Models (AMMI) are widely used to analyze main effects and genotype by environment (GEN. number=TRUE.freq Author(s) Felipe de Mendiburu Maintainer: Felipe de Mendiburu <f. ENV) interactions in multilocation variety trials. GEN. Furthermore. plot graphics parameters Details additional biplot.

5e-8)) .5e-8.crepress.AMMI(plrv[.data2) attach(cic) #startgraph par(cex=0. plrv[. plrv[..AMMI(plrv[.5]. plrv[.2].5]. plrv[.2].graph="biplot") model<. plrv[.c(1.19.5].18)] data2<-CIC$oxapampa[.AMMI(plrv[. Rep.graph="triplot".6. geneticists.20)] cic <.2]. Genotype. plrv[.1.graph="triplot") model<.7. www. plrv[.3].6) model<-AMMI(Locality.17.5]. Weikai Yan and Manjit S.number=FALSE) # triplot model<.3].AMMI(plrv[. relative. de Mendiburu References GGE Biplot Analysis: A graphical tool for breeder. plrv[.3]. plrv[.1]. others parameters Author(s) F.graph="biplot".c(1.6.1].6 AMMI Value ENV Factor GEN Factor REP Numeric Y Numeric MSE Numeric number TRUE or FALSE graph "biplot" or "triplot" .1].1]. Kang.7. plrv[. and agronomists. plrv[. Principles and procedures of statistics: a biometrical approach Steel & Torry & Dickey.com 2003. ylim=c(-1.. Third Edition 1997 See Also lineXtester Examples # Full replications library(agricolae) library(klaR) # Example 1 data(plrv) #startgraph # biplot model<.rbind(data1.2].3].number=FALSE) #endgraph # Example 2 data(CIC) data1<-CIC$comas[. plrv[.

YLD. MSerror) #endgraph pc<.PC1.labels=row.11 0.contour 7 #endgraph pc<. Mean square error is well-known. data(sinRepAmmi) attach(sinRepAmmi) REP <.3 MSerror <.seq(length(YLD)) arrows(MEANS. main="AMMI BIPLOT".cex=0. col= "brown".8. 0.lty=2.1:4] attach(bplot) #startgraph par(cex=0.contour AMMI contour . cor = FALSE) pc$loadings summary(pc) model$biplot detach(cic) # Example 3 # Only means.col="red") amb<-subset(bplot. GEN.length=0.93.24224 #startgraph model<-AMMI(ENV. 0. lwd=1.PC1.col="blue").type=="ENV") detach(bplot) attach(amb) s <. cor = FALSE) pc$loadings summary(pc) model$biplot detach(sinRepAmmi) # Biplot with the one restored observed. rm(REP) bplot<-model$biplot[.50 3. REP.0.58 13.1) #endgraph detach(amb) # Principal components by means of the covariance # It is to compare results with AMMI cova<-cov(model$genXenv) values<-eigen(cova) total<-sum(values$values) round(values$values*100/total.9*(YLD[s]-MEANS)+MEANS. 0.AMMI.v= MEANS.8) plot(YLD.princomp(model$genXenv.frame=TRUE) MEANS<-mean(YLD) abline(h=0.9*PC1[s].princomp(model$genXenv.00 AMMI.81 18.names(bplot).2) # AMMI: 64.text(YLD.

8

AMMI.contour

Description
Draws a polygon or a circumference around the center of the Biplot with a proportional radio at the
longest distance of the genotype.
Usage
AMMI.contour(model, distance, shape, ...)
Arguments
model

Object

distance

Circumference radius >0 and <=1

shape

Numerical, relating to the shape of the polygon outline.

...

Parameters corresponding to the R lines function

Details
First, it is necessary to execute the AMMI function. It is only valid for the BIPLOT function but
not for the TRIPLOT one.
Value
model

output AMMI

distance

Numeric >0 and <=1

shape

Numeric

Note
Complement graphics AMMI
Author(s)
Felipe de Mendiburu
See Also
AMMI
Examples
library(agricolae)
# see AMMI.
data(sinRepAmmi)
attach(sinRepAmmi)
REP <- 3
MSerror <- 93.24224
# Example 1
#startgraph
model<-AMMI(ENV, GEN, REP, YLD, MSerror)

audpc

9

AMMI.contour(model,distance=0.7,shape=8,col="red",lwd=2,lty=5)
#endgraph
# Example 2
#startgraph
for (i in seq(0,0.7,length=15)) {
AMMI.contour(model,distance=i,shape=100,col="green",lwd=2)
}

Calculating the absolute or relative value of the AUDPC

audpc

Description
Area Under Disease Progress Curve. The AUDPC measures the disease throughout a period. The
AUDPC is the area that is determined by the sum of trapezes under the curve.
Usage
audpc(evaluation, dates, type = "absolute")
Arguments
evaluation

Table of data of the evaluations: Data frame

dates

Vector of dates corresponding to each evaluation

type

relative, absolute

Details
AUDPC. For the illustration one considers three evaluations (14, 21 and 28 days) and percentage of
damage in the plant 40, 80 and 90 (interval between dates of evaluation 7 days). AUDPC = 1045.
The evaluations can be at different interval.
Value
evaluation

data frame

dates

a numeric vector

type

text

Author(s)
Felipe de Mendiburu
References
Campbell, C. L., L. V. Madden. (1990): Introduction to Plant Disease Epidemiology. John Wiley
& Sons, New York City.

10

bar.err

Examples
library(agricolae)
# example 1
dates<-c(14,21,28) # days
evaluation<-data.frame(E1=40,E2=80,E3=90) # percentages
plot(dates,evaluation,type="h",ylim=c(0,100),col="red",axes=FALSE)
title(cex.main=0.8,main="Absolute or Relative AUDPC\nTotal area = 100*(28-14)=1400")
lines(dates,evaluation,col="red")
text(dates,evaluation+5,evaluation)
text(18,20,"A = (21-14)*(80+40)/2")
text(25,60,"B = (28-21)*(90+80)/2")
text(25,40,"audpc = A+B = 1015")
text(24.5,33,"relative = audpc/area = 0.725")
abline(h=0)
axis(1,dates)
axis(2,seq(0,100,5),las=2)
lines(rbind(c(14,40),c(14,100)),lty=8,col="green")
lines(rbind(c(14,100),c(28,100)),lty=8,col="green")
lines(rbind(c(28,90),c(28,100)),lty=8,col="green")
# It calculates audpc absolute
absolute<-audpc(evaluation,dates,type="absolute")
print(absolute)
rm(evaluation, dates, absolute)
# example 2
data(disease)
dates<-c(1,2,3) # week
evaluation<-disease[,c(4,5,6)]
# It calculates audpc relative
index <-audpc(evaluation, dates, type = "relative")
# Correlation between the yield and audpc
correlation(disease$yield, index, method="kendall")
# example 3
data(CIC)
comas <- CIC$comas
oxapampa <- CIC$oxapampa
dcomas <- names(comas)[9:16]
days<- as.numeric(substr(dcomas,2,3))
AUDPC<- audpc(comas[,9:16],days)
relative<-audpc(comas[,9:16],days,type = "relative")
h1<-graph.freq(AUDPC,border="red",density=4,col="blue")
table.freq(h1)
h2<-graph.freq(relative,border="red",density=4,col="blue",
frequency=2, ylab="relative frequency")

bar.err

Plotting the standard error or standard deviance of a multiple comparison of means

waller.virus. It uses the objects generated by a procedure of comparison like LSD. . HSD.. main="Yield of sweetpotato\ndealt with different virus") detach(sweetpotato) # std = FALSE (default) is standard error #startgraph . MSerror.etc std Standard deviation or standar error horiz Horizontal or vertical bars .data=sweetpotato) df<-df..bar. Kruskal and Waller-Duncan.. Fc. Usage bar..test..group Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus. HSD.test. std = TRUE. std = T : standard deviation.. Parameters of the function barplot() Details x: data frame formed by 5 columns: name of the bars.residual(model) MSerror<-deviance(model)/df Fc<-anova(model)[1.4] attach(sweetpotato) comparison <. height.err(x. bar. Value x object std TRUE or FALSE horiz TRUE or FALSE . horiz = FALSE. std = F : standard error. waller..test(yield.. kruskal. level replications and standard error of the bar.test.df.. Parameters of the function barplot() Author(s) Felipe de Mendiburu See Also LSD.) Arguments x object or comparisons the LSD.test..test.err 11 Description It plots bars of the averages of treatments and standard error or standard deviance. HSD.

It can also display the ’average’ value over each bar in a bar chart.col="green"..2).45)...err(comparison.density=4.density=10. Value x Data frame horiz Logical TRUE or FALSE .test. HSD. main="Standard deviation") bar. main="Standard error") par(mfrow=c(1. It uses the objects generated by a procedure of comparison like LSD. horiz = FALSE. Parameters of the function barplot() Details x: data frame formed by 5 columns: name of the bars.col="blue".1)) #endgraph Plotting the multiple comparison of means bar. height and level of the bar. col="brown".test .test.err(comparison.45).err(comparison. Friedman or Durbin..std=FALSE.density=6.horiz=TRUE. waller.test.ylim=c(0.xlim=c(0.. kruskal . .45). Kruskall.group Description It plots bars of the averages of treatments to compare.group(x.ylim=c(0. Waller-Duncan..12 bar. friedman. durbin.angle=125.horiz=TRUE.angle=125.45).) Arguments x Object created by a test of comparison horiz Horizontal or vertical bars .cex=1) bar.xlim=c(0.density=8. Usage bar.std=FALSE.main="Standard error") bar. main="Standard deviation") bar.err(comparison.group par(mfrow=c(2. Author(s) Felipe de Meniburu See Also LSD. HSD.

Usage BIB. 0. 0.test(yield.group(xy.44.alpha=0.29.border=FALSE.xlim=c(0.MSerror.09.las=1) #endgraph Finding the Variance Analysis of the Balanced Incomplete Block Design BIB.frame(x.group(comparison.test(block. ylab="Virus") #endgraph # Example 2 library(agricolae) x <.y.49) xy <.test Description Analysis of variance BIB and comparison mean adjusted. method = c("lsd".density=30.main=0.LSD.test 13 Examples # Example 1 library(agricolae) data(sweetpotato) model<-aov(yield~virus.ylim=c(0."tukey".50)) title(cex.group=TRUE) detach(sweetpotato) #startgraph par(cex=1."duncan".horiz=TRUE.01.col="brown".BIB.5) bar.lwd=2. y. 0.data.col="blue".c(0.6).df.1:4 y <.5) bar.border="red".residual(model) MSerror<-deviance(model)/df attach(sweetpotato) comparison<.main="Comparison between\ntreatment means".density=8. alpha = Arguments block blocks trt Treatment y Response method Comparison treatments alpha Significant test group logical .data=sweetpotato) df<-df."snk").angle=90. trt. virus.0.8.y) #startgraph par(cex=1."waller".xlab="Yield".

monovinyl.37) out<-BIB. Kuehl.45.3. India # 6 varieties of wheat crop in a BIB whit 10 blocks of 3 plots each.4.method="tukey".67.18.60)) # Example linear estimation and design of experiments.8.550.60.54.45. duncan: Duncan’s new multiple range test waller: Waller-Duncan test.21.475.26.61.4.c(1. Joshi.550.test(run.63. lsd: Least significant difference.59. D.400.8.400.65.10) .test Details Method of comparison treatment.5. Duxbury.65.33.21. tukey: Honestly significant differente.D.7.35. 250.20.39.475) monovinyl<-c(16.1.400.9.4.38. 2000 Linear Estimation and Design of Experiments.31.Hill Book Company.6. 2001 run<-gl(10. Joshi. New Delhi.70.32.52.475.19.13.ylim=c(0. Edicion. India.19.24.68.77.test(run.475.55.325.13.250.group=TRUE) out<-BIB.72.density=4.61.57.250. snk: StudentNewman-Keuls (SNK) Value block Vector trt Vector y numeric vector method Character alpha Numeric group TRUE or FALSE Author(s) F.75.550. 55.2.10.err(out.63.5.5.2.56) varieties<-gl(6.2. Robert O.3.46.8.10.6.test(run.7.65. Ching Chun Li McGraw .D. Inc.psi.60.400.400.test(run.250.50.24.method="waller".3) psi<-c(250.method="waller".65..325.62.group=FALSE) bar.7.59.50.test Examples library(agricolae) # Example Design of Experiments.475.475.3. Institute of Social Sciences Agra.550. D.325.group=TRUE) out<-BIB.325. Introduction to experimental statistics.550.47. 2da. 2nd ed.monovinyl.250.psi. Robert O.54.6.monovinyl.34.10.group=FALSE) out<-BIB. 30. 1964 See Also durbin.9.62.psi.325.48.400. 1987 # Professor of Statistics. Kuehl. y <-c(69.325. de Mendiburu References Design of Experiments.1.psi.550.9.monovinyl. WILEY EASTERN LIMITED 1987.54.39.method="tukey".31.14 BIB.5) block <. New York.

col="blue".14.group(model.1. y <-c(10. 1979 .4. Value model 1. replication.2.5.2 and 3.3) trt <.6.test(block. female.2.5.5.4.4. male. y) # Example Introduction to experimental statistics. response . carolina 2 : set.17. replication.19. Author(s) Felipe de Mendiburu References Biometrical Methods in Quantitative Genetic Analysis. response .. Usage carolina(model. male.15. female.21) block<-gl(7.c(1.7) model<-BIB.3.4. Chaudhary.density=4. II and III genetic designs. varieties. Singh.. trt.3. replication.2 # 7 trt.. 1964 # pag.10.ylim=c(0.6.test(block..2. k=3 and b=7. 27.7. 2 or 3 data in order carolina 1 : set. male.14.14.max(y))) North Carolina Designs I.12.6.8.7. response . y. progeny.1.5. method="duncan") bar. II and III carolina Description Statistic analysis of the Carolina I. 395 table.12..carolina 15 BIB.. female. carolina 3 : set.6.data) Arguments model Constant data Data frame Details model = 1.3.15.11.2 and 3 is I. Ching Chun Li.19. II and III see carolina1.10.5.11.

c(2.4.16 Chz2006 See Also DC Examples library(agricolae) data(DC) carolina1 <.3. wilt percentage at 60 days health a numeric vector. Usage data(Chz2006) Format The format is: List of 2 amendment a factor crop a factor block a numeric vector. kg/8m2 .DC$carolina2 # str(carolina2) majes<-subset(carolina2.1]==1) majes<-majes[. kg/8m2 rot a numeric vector. after application of inorganic amendments and a rotation crop in Carhuaz Peru. number plant wilt_percent a numeric vector.carolina3) output[][-1] Chz2006 Data amendment Carhuaz 2006 Description Incidents and performance of healthy tubers and rotten potato field infested with naturally Ralstonia solanacearum Race 3/Bv 2A. replications plant a numeric vector.carolina1) output[][-1] carolina2 <.carolina2[.DC$carolina3 # str(carolina3) output<-carolina(model=3.majes[.c(1:4. 2006.DC$carolina1 # str(carolina1) output<-carolina(model=1.5.6:8)] output<-carolina(model=2.6)]) output[][-1] carolina3 <.

1:3].wilt[.function(x) mean(x. Source Experimental field.yield[.method="spearman") CIC Data for late blight of potatoes Description A study of Phytophthora infestans in the potato plant in the localities of Comas and Oxapampa in Peru.model(model) yield<-yield[order(paste(yield[. replications E9 a numeric vector. relative area . oxapampa) Locality a factor with levels Comas Oxapampa Genotype a factor Rep a numeric vector.] correlation(means[.stat(wilt[. Examples library(agricolae) data(Chz2006) str(Chz2006) wilt<-Chz2006$wilt yield<-Chz2006$yield means <. Usage data(CIC) Format The format is: List of 2 (comas.2].5].aov(wilt_percent ~ block + crop.4]. References International Potato Center.CIC 17 Details Application of inorganic amendment and crop rotation to control bacterial wilt of the potato (MBP).tapply.Lima Peru. means) anova(model) cv.4].na.1]. infestans percentaje to 9 days AUDPC a numeric vector: the area under the disease-progress curve Relative a numeric vector.3])). 2005. 2006.yield[. Data Kindly provided by Pedro Aley.rm=TRUE)) names(means)[4]<-"wilt_percent" model <. CIP .yield[.

Lima Peru. Data Kindly provided by Matilde Orrillo. 2004-2005. per. help(audpc) Exx: Evaluation in percentaje. Source Experimental field.Lima Peru. ORD2. Usage data(clay) Format A data frame with 69 observations on the following 3 variables. .9 Fahrenheit.3 oxapampa: temperature=64.clay a numeric vector days a numeric vector ralstonia a numeric vector Source Experimental field. CIP .2 AUDPC and relative see function audpc(). ORD1. SBLK and row are references location of the plot in the field. relative humidity=86. Examples library(agricolae) data(CIC) CIC$comas CIC$oxapampa clay Data of Ralstonia population in clay soil Description An evaluation over a time period.18 clay Details comas: temperature=59. relative humidity=83. CIP . References International Potato Center.8 Fahrenheit. xx is days. References International Potato Center.

Spreader rows around plots were used at each site. 4. rAUDPC rankings were analyzed for phenotypic stability with nonparametric measures. cultivar a factor with 56 levels replication a factor with 3 levels comas a numeric vector oxapampa a numeric vector Details The experimental design was a randomized complete block design with 3 replications of 15 apical stem cuttings in Oxapampa and 10 tubers in Mariscal Castilla.Oxapampa Description Fifty-three potato varieties developed by the breeding program of the International Potato Center and released in different countries around the world were evaluated for their resistance to late blight in two locations in Peru. then values were converted to the relative area under the diseases progress curve (rAUPDC). 11) at a concentration of 5000-sporangia/ ml at 49 days after planting. Data Kindly provided by Wilmer Perez. Source Experimental field. after which the plants were left to natural infection. 2002.9 m between-row spacings. Plots were 11. 10. Mancozeb was applied weekly until 30 days after transplanting or planting. with virulence R1. 3. 2. inoculum was enhanced with local isolate (POX 067. Percentage of foliar infection was estimated visually every 3 days for 8 times in Oxapampa and every 7 days for 12 times in Comas. 7. . References International Potato Center. Usage data(ComasOxapampa) Format A data frame with 168 observations on the following 4 variables.5 m in size with 30 cm in-row and 0.ComasOxapampa 19 Examples library(agricolae) data(clay) str(clay) ComasOxapampa Data AUDPC Comas . Due to climatic conditions not conductive to the disease in Oxapampa. CIP . 6.9 x 18. 5.Lima Peru.

col. Mariscal Castilla (11 42 54 S latitude. method="kendall") consensus consensus of clusters Description The criterion of the consensus is to produce many trees by means of boostrap and to such calculate the relative frequency with members of the clusters. "mcquitty".) Arguments data data frame distance method distance.l.cex."maximum". "centroid". .s. Cruza 148 (moderately resistant) and Pimpernell (susceptible str(ComasOxapampa) means <.mean) correlation(means$comas.text size text on percentage consensus col. hclust() and daisy() of cluster.20 consensus Examples library(agricolae) data(ComasOxapampa) # Oxapampa (10 35 31 S latitude. "canberra".. see dist() method method cluster. 2800 m."average". "minkowski".text=1. "single". "minkowski". Method: "ward". see hclust() nboot The number of bootstrap samples desired. "gower"). Usage consensus(data..a.a.means$oxapampa. 75 23 0 E longitude. "manhattan". "centroid"). "complete".text color text on percentage consensus . "maximum". "binary".3:4].tapply. 1813 m. duplicate control is TRUE other case is FALSE cex.text="red". "average".distance=c("binary".."median". "canberra". "mcquitty".l ) # Comas."manhattan". .stat(ComasOxapampa[. "median".duplicate=TRUE.s.) # cultivars LBr-40 (resistant).nboot=500."euclidean". see functions: dist().ComasOxapampa$cultivar."single".."ward". "gower". 75 04 45 E longitude.method=c("complete". parameters of the plot dendrogram Details distance: "euclidean".

data[.3].5)] Groups<-Groups[order(Groups[.2].nboot=100) # Order consensus Groups<-output$table.data[. Tibshirani.edgePar = list(lty=1:2.6) par(cex=0.3]. de Mendiburu References An Introduction to the Boostrap.dend plot(dend) text(data[. col=2:1)) text(data[.type="r".col="blue". col=2:1)) text(data[.col="blue".decreasing=TRUE).c(6. the rownames is necesary’ nboot integer duplicate logical TRUE or FALSE cex.data[. method="complete". cbind(output$dendrogram$labels) # To reproduce dendrogram dend<-output$dendrogram data<-output$table.edgePar = list(lty=1:2.5].8) output<-consensus( pamCIP.5]) # Other examples # classical dendrogram dend<-as.consensus 21 Value data numerical.text color text on consensus Author(s) F.4].duplicate=FALSE.3].cex=1) # Without the control of duplicates output<-consensus( pamCIP.dend[.5]. Bradley Efron and Robert J. Chapman and Hall/CRC See Also hclust.1.] print(Groups) # Identification of the codes with the numbers.dendrogram(output$dendrogram) plot(dend.nboot=100) .data[.type="t". hgroups.text size text on consensus col.data[.data[.4]. 1993.4].cex=1) # plot(dend. hcut Examples library(agricolae) data(pamCIP) # only code rownames(pamCIP)<-substr(rownames(pamCIP).distance="binary".

Usage data(corn) Format A data frame with 34 observations on the following 3 variables. Ordinarily. Conover. Ties values receive the averarge rank. method a numeric vector observation a numeric vector rx a numeric vector Details The observations are ranked from the smallest.J. Examples data(corn) str(corn) .22 corn # using distance gower library(cluster) output<-consensus( pamCIP. of rank 1 to the largest 101.nboot=100) corn Data of corn Description Data from a completely randomized design where four different methods of growing corn resulted in various yields per acre on various plots of ground where the four methods were tried. Third Edition. of rank N=34. Source Book: Practical Nonparametric Statistics. References Practical Nonparametrics Statistics. method="complete". 77. W.distance="gower". 1999. but here we will use the kuskal-wallis test so that a rough comparison may be made with the mediasn test. only one statistical analysis is used.

clay. Methods of Kendall. Spearman and Pearson.alternative="two.sided") Arguments x Vector y Vector method "pearson".clay. Usage correl(x. Author(s) Felipe de Mendiburu References Numerical Recipes in C. "kendall".. y. method = "pearson".method="pearson") . Second Edition.method="spearman") correl(pH. "greater" Value x Numeric y Numeric . "spearman" alternative "two. See Also correlation Examples library(agricolae) data(soil) attach(soil) correl(pH.correl 23 Correlation Coefficient correl Description An exact correlation for ties or without ties.sided".method="kendall") correl(pH. "less".clay..

sided") Arguments x table. The methods to apply are Pearson.sided". method = c("pearson".24 correlation Correlation analysis. The results are similar to SAS. "spearman" alternative "two. Usage correlation(x. "spearman") . "kendall".alternative="two. data(soil) attach(soil) . the Pearson method will be used. variable 2 and other elements from soil.method="pearson") analysis # Example 2: correlation between pH. Spearman and Kendall. "kendall". Spearman and Kendall correlation Description It obtains the coefficients of correlation and p-value between all the variables of a data table.y=NULL. matrix or vector method "pearson". "greater" Details Parameters equal to function cor() Value table Numeric Author(s) Felipe de Mendiburu See Also correl Examples library(agricolae) # example 1 data(soil) analysis<-correlation(soil[. matrix or vector y table.2:8]. "less". Methods of Pearson. In case of not specifying the method.

Autor: Calzada Benza Universidad Nacional Agraria .Peru.soil[.cotton 25 analysis<-correlation(pH. Examples library(agricolae) data(cotton) str(cotton) .method="pearson". Autor: Calzada Benza Universidad Nacional Agraria .sided") detach(soil) cotton Data of cotton Description Data of cotton collected in experiments of two localities in Lima and Pisco.clay. References Book spanish: Metodos estadisticos para la investigacion.Peru. alternative="two.3:8]. data(soil) attach(soil) correlation(pH..La Molina . site a factor with levels Lima Pisco block a factor with levels I II III IV V VI lineage a numeric vector epoca a numeric vector yield a numeric vector Source Book spanish: Metodos estadisticos para la investigacion.La Molina . Usage data(cotton) Format A data frame with 96 observations on the following 5 variables.method="kendall".alternative="less") analysis detach(soil) # Example 3: correlation between pH and clay method kendall. Peru.

model(model) . Waller-Duncan or HSD and complete with it: library(agricolae) # not run # cv<-cv.test Examples # see examples from LSD . waller.test.model Description It obtains the coefficient of variation of the experiment obtained by models lm() or aov() Usage cv. HSD.test.26 cv.model Coefficient of the experiment variation cv.model(x) Arguments x object of model lm() or AOV() Details sqrt(MSerror)*100/mean(x) Value x object Author(s) Felipe de Mendiburu See Also LSD.

similarity Description This process consists of finding the coefficient of the distances of similarity of binary tables (1 and 0) as used for scoring molecular marker data for presence and absence of PCR amplification products. Usage cv.similarity 27 Coefficient of the similarity matrix variation cv.cv Examples # molecular markers. resampling.similarity(markers) .cv. library(agricolae) data(markers) cv<-cv.similarity(A) Arguments A matrix of binary data Value A Data frame or matrix. numerics Author(s) Felipe de Mendiburu See Also similarity.

HSD.test(block. tukey: Honestly significant differente. trt. Usage DAU. PBIB. W. method = c("lsd". T.28 DAU. lsd: Least significant difference. See Also BIB. Value block Vector trt Vector y numeric vector method Character Author(s) F. Record LV(2):191208.test Finding the Variance Analysis of the Augmented block Design DAU. y. (1956).LSD.test Description Analysis of variance Augmented block and comparison mean adjusted. de Mendiburu References Federer."tukey")) Arguments block blocks trt Treatment y Response method Comparison treatments Details Method of comparison treatment.test.test.test .test. Augmented (or hoonuiaku) designs. Hawaiian Planters.

A set of 4 randomly selected plans are used as females and are mated to the above male. No female is used for any second mating. 1979.78.96. is taken as base population.79.77.test(block. 3 data. Thus n1 x n2 progenies are produced whitch are analysed in a suitably laid experiment."A". n1 males and n2 females are randomly selected and each male is crossed to each of the females.7)) trt<-c("A". References Biometrical Methods in Quantitative Genetic Analysis. 9 var) and carolina3(64 obs. Similarly."k".method="lsd") DC Data for the analysis of carolina genetic design Description Data for the analysis of carolina I.74."B". From the population an individual is randomly selected and used as a male. Thus a set of 4 full-sib families are produced. Singh. Chaudhary. Chaudhary."i". Both paternal and maternal half-sibs are produced in this design.70.78.82) model<."C".81. 6 var).78. carolina2(300 obs. Carolina1: Data for the analysis of Carolina I Genetic design.91. .79.81."C".DC 29 Examples library(agricolae) block<-c(rep("I"."g".7). a large number of male groups are produced. 1979."h"."f".6). say L1 and L2.rep("II".89.75.79. Carolina2: Data for the analysis of Carolina II Genetic design.frame: carolina1(72 obs."B".81.92. II and III genetic design Usage data(DC) Details DC is list."D". This is denoted as a male group. four male groups (16 female groups) from a set.87."l"."A".DAU.yield. Singh. In this design F2 or any advanced generation maintained by random mating."D". The material for estimation of genetic parameters is produced by back crossing randomly selected F2 individuals (using as males) to each of the inbreds (used as females)."B". From an F2 population."D"."C". Carolina3: Data for the analysis of Carolina III genetic design. 5 var). produced from cross between two pure-lines.rep("III". Source Biometrical Methods in Quantitative Genetic Analysis.trt."j") yield<-c(83."e". The F2 population is produced by crossing two inbreds.

na(x.] 3 .5.1] #[1.0.1] [.NA) dim(x)<-c(4.7.5.] 7 0 NA delete.30 delete.2] [. "greater") ) Arguments x matrix with NA alternative "less" or "greater" Value x matrix Author(s) Felipe de Mendiburu Examples library(agricolae) x<-c(2.3.3. Usage delete.8.3) x # [.3] #[1. alternative=c("less".na Examples data(DC) names(DC) str(DC$carolina1) str(DC$carolina2) str(DC$carolina3) Omitting the rows or columns with missing observations of a matrix (NA) delete.] 2 #[2.na Description In many situations it is required to omit the rows or columns less or greater with NA of the matrix.na(x."less") # [.NA.] 2 5 4 #[2.] 5 #[3.4.NA.] 5 NA 3 #[3.] 3 8 NA #[4.

] 31 7 delete.design.1] [. "default" ) Value A vector. number = 1. B.ab #[4. "Marsaglia-Multicarry". The seed is by set. kinds). "user-supplied".] 2 5 4 Design in blocks for factorial pxq design. "Knuth-TAOCP-2002".na(x."greater") # [.3] #[1. seed = 0. name of the levels of the second factor r Numeric number Numeric seed Numeric Author(s) Felipe de Mendiburu .2] [. "KnuthTAOCP". "Random" uses the methods of number generation in R.ab(A. r. name of the levels of the first factor B vector. "Mersenne-Twister". Usage design.c("Wichmann-Hill".seed(seed. kinds = "Super-Duper") Arguments A Levels of A B Levels of B r Replications or Blocks number Number of first plot seed Seed kinds Method for to randomize Details kinds <.ab Description It generates a design of blocks for combined pxq. "Super-Duper".

ab <-design. Ching Chun Li. IV."canchan".c("perricholi". fact. II. r.alpha(trt.80. McGraw-Hill Book Company.r=3. kg/ha library(agricolae) variety <. seed = 0.120) # level of nitrogen rcbd. s odd but not a multiple of 3. kinds = "Super-Duper") Arguments trt Treatments k size block r Replications number number of first plot seed seed kinds method for to randomize Details Series: I.rcbd. s even.crd. number=1001) print(rcbd. r=4. III. design.alpha Description Creates alpha designs starting from the alpha design fixing under the 4 series formulated by Patterson and Williams. k <= s-1.lsd. design. k<=s r= replications s=number of blocks k=size of block Number of treatment is equal to k*s . r=3. number = 1. k. nitrogen.ab. k <= s.c(40."tomasa") nitrogen <.ab) # print of the field book Alpha design type (0. k <= s. INC.ab(variety.1) design. Usage design. s odd.alpha References Introduction to Experimental Statistics.32 design. 5. r=2. York. New.nk Examples # factorial 3 x 3 with 5 replications # 3 varieties from potato and nitrogen in the fertilization. random. 1964 See Also design.

numeric character Author(s) Felipe de Mendiburu References H. Patterson and E.. printed in Great Britain.design.r) for (i in 1:r) print(t(plots[.k.i])) book$design # Example two trt<-letters[1:12] t <.r) plots<-book$book[.org/cgi/content/abstract/63/1/83 See Also design.design. name of the treatments Constant.s.s.r) plots<-book$book[.lattice.alpha 33 Value trt k r number seed kinds vector.oxfordjournals. numeric Constant.cyclic Examples library(agricolae) #Example one trt<-1:30 t <. Online: http://biomet.length(trt) k<-3 r<-3 s<-t/k book<. Williams.2 book<.i])) book$design book$alpha .alpha(trt.R.D.1] dim(plots)<-c(k. Biometrika (1976) A new class of resolvable incomplete block designs.design.alpha(trt.. design.k.bib.1] dim(plots)<-c(k.r) for (i in 1:r) print(t(plots[. design. numeric Constant.length(trt) # size block k k<-3 # Blocks s s<-t/k # replications r r <. numeric Constant.

seed(seed. kinds). Robert O. k. design. "KnuthTAOCP". "Mersenne-Twister".bib Description Creates Randomized Balanced Incomplete Block Design..lsd. seed = 0.rcbd .nk . "default" ) Value trt vector. "Random" uses the methods of number generation in R.bib(trt.bib Randomized Balanced Incomplete Block Designs. kinds = "Super-Duper") Arguments trt Treatments k size block number number of first plot seed seed kinds method for to randomize Details kinds <.fact. Kuehl. number = 1. name of the treatments k number. 2000 See Also design. random. "Super-Duper". "user-supplied". The seed is by set. Duxbury.c("Wichmann-Hill". numeric number Numeric seed Numeric kinds character Author(s) Felipe de Mendiburu References Design of Experiments. design. Usage design. BIB design. "Knuth-TAOCP-2002".crd. 2nd ed.ab.34 design. "Marsaglia-Multicarry".

3]) t(matrix(plots. numeric r Numeric number Numeric seed Numeric .c(3.kinds ="Super-Duper") # seed = 41 plots <-as. r."D") k<-3 bib <-design.crd 35 Examples library(agricolae) # 4 treatments and k=3 size block trt<-c("A".character(bib[.numeric(bib[.number=101. "default" ) Value trt vector.seed =41.show("bib. name of the treatments r vector.k. "user-supplied". kinds).design. number = 1.c(3.4))) t(matrix(field. "KnuthTAOCP"."B".1]) field <-as. sep="\t") # file. "Marsaglia-Multicarry".bib(trt. The seed is by set."bib. kinds = "Super-Duper") Arguments trt Treatments r Replications number number of first plot seed seed kinds method for to randomize Details kinds <.txt". Usage design.txt") Completely Randomized Design design.table(bib.c("Wichmann-Hill".crd Description It generates completely a randomized design with equal or different repetition. "Super-Duper". row.names=FALSE."C". "Mersenne-Twister". "Random" uses the methods of number generation in R. seed = 0.4))) # write in hard disk # write.seed(seed. "Knuth-TAOCP-2002".crd(trt.

36 design.12543."CIP-401". design. New.nk Examples library(agricolae) cipnumber <-c("CIP-101".cyclic Author(s) Felipe de Mendiburu References Introduction to Experimental Statistics."CIP-301"."crd. McGraw-Hill Book Company. Usage design.lsd. The efficient and robust cyclic designs for 6 to 30 treatments."CIP-501") rep <-c(4.3."Mersenne-Twister") # no seed crd2 <-design.number=101) # write to hard disk # write. row. fact."CIP-201".crd(cipnumber. k.table(crd1. York. design. replications <= 10. kinds = "Super-Duper Arguments trt k r number rowcol seed kinds vector treatments block size Replications number of first plot TRUE: row-column design init seed random random method .show("crd.ab.cyclic Cyclic designs Description The cyclic design is a incomplete blocks designs.names=FALSE. the plan is generated and randomized.rep.cyclic(trt.crd(cipnumber. it is generated from a incomplete block initial of the size k.rcbd.4. number = 1.txt") design.rep. Ching Chun Li. INC.5.number=101. seed = 0. sep="\t") # file. 1964 See Also design.3) # seed = 12543 crd1 <-design. rowcol = FALSE.txt". r.

43. Duxbury. replication = 6 plan1 <. Robert(2000).k=2.2.design. numeric seed Constant. replication = i*k. 1.cyclic 37 Details Number o treatment 6 to 30.lattice.design. name of the treatments k Constant.design. R. "i" is value integer. Design of Experiments.alpha Examples library(agricolae) trt<-letters[1:8] # block size = 2. Soc.A.. (1981) Efficient Cyclic Design. See Also design.k=2. J. No. 2nd ed. (r) Replication 2 to 10. Value trt vector.3 plan1$design[[1]] plan1$design[[2]] plan1$design[[3]] plan1$book # row-column design plan2 <. numeric kinds character Author(s) Felipe de Mendiburu References Kuehl. B. Statist. design. r=6. design.cyclic(trt. J. rowcol=TRUE) plan2$design .bib. (k) size of block 2 to 10. r=6) names(plan1) # groups 1. John. pp.cyclic(trt. numeric number Constant. 76-80.

"Super-Duper".38 design.. vol. "KnuthTAOCP". Augmented (or Hoonuiaku) Design.dau Augmented block design design. r. Federer. (1956)..c("Wichmann-Hill". 29-35 . (1975). In Augmented Designs. pp. some treatments (checks) are replicate r times and other treatments (new) are replicated once. "user-supplied". 1 (mar. Bometrics.dau Description In these designs. 55: 191-208. No. rec. Hawaii Plr. 2.T.T and Raghavarao. "Marsaglia-Multicarry". name of the treatments r Numeric number Numeric seed Numeric Author(s) Felipe de Mendiburu References 1. trt2. Usage design. seed = 0. 31. number = 1. kinds = "Super-Duper". Federer. name of the treatments trt2 vector. "Knuth-TAOCP-2002". W. 1975). name="trt") Arguments trt1 checks trt2 new r Replications or blocks number number of first plot seed seed kinds method for to randomize name name of treatments Details kinds <. D. "Mersenne-Twister".dau(trt1. W. "default" ) Value trt1 vector..

1) book <."-". sep="\t") # file.c("Wichmann-Hill".as. each at k levels.4."D") T2<-letters[20:26] dau <-design."C".rcbd Examples library(agricolae) # 4 treatments and 5 blocks T1<-c("A".latin square is a KxK pattern that permits the study of k treatments simultaneously with three different blocking variables. kinds = "Super-Duper") Arguments trt1 Treatments trt2 Treatments number number of first plot seed seed kinds method for to randomize Details kinds <. seed = 0.txt") # Augmented designs in Completely Randomized Design trt<-c(T1. r=5) # field book dau by(dau.latin square design design.function(x) paste(x[.1.1].graeco(trt1.names=FALSE.design. 8.r) Graeco .1.1."B".character(x[.3]))) # write in hard disk # write. "Knuth-TAOCP-2002"."dau. "Marsaglia-Multicarry".dau(T1.graeco 39 See Also design.T2.crd(trt. "Mersenne-Twister".dau[2].txt".table(dau.1.4.4.1. number = 1. "KnuthTAOCP".1.design. "Super-Duper".show("dau.T2) r<-c(4. row. "user-supplied". 10 and 12) Usage design. The function is only for squares of the odd numbers and even numbers (4.graeco Description A graeco . "default" ) . trt2.

"x".1]) trt <.4]. random.paste(graeco[.10) print(t(trt)) . 2005.graeco(T1."w"."d") T2<-c("v".graeco[. fact.graeco[. Innovation. See Also design. and Discovery Second Edition.numeric(graeco[. Experimental design.letters[1:10] T2 <. name of the treatments trt2 vector.crd.lsd. Statistics for Experimenters Design. Wiley-Interscience.number=101) plots <-as.4) dim(trt) <-c(4.graeco Value trt1 vector.design.ab."b".4) print(t(plots)) print(t(trt)) # 10 x 10 T1 <. George E."c".1:10 graeco <.5]) dim(trt) <-c(10. name of the treatments number Numeric seed Numeric Author(s) Felipe de Mendiburu References 1. Box. 2.T2) trt <.paste(graeco[. P."y") graeco <.40 design.graeco(T1.nk Examples library(agricolae) T1<-c("a". design.T2. Cochran and Cox.design.4]. Second edition.5]) dim(plots)<-c(4. Wiley Classics Library Edition published 1992.

"user-supplied".crd.. "Mersenne-Twister".lattice 41 Lattice designs design. Erwin L. 1962.lattice(3. fact.ab. "Super-Duper". It randomizes treatments in k x k lattice. Usage design. "KnuthTAOCP".type="simple") # simple lattice desing. LeCLERG. design.lattice(10. numeric Numeric Numeric Author(s) Felipe de Mendiburu References FIELD PLOT TECHNIQUE.rcbd. type="triple". "Knuth-TAOCP-2002".type="triple") # triple lattice design.lattice(k. random.lattice Description SIMPLE and TRIPLE lattice designs. Minnesota See Also design. Burgess Publishing Company. 3x3 ( 9 varieties) . "default" ) Value k number seed vector. number = 1. 10x10 (100 varieties) design. kinds = "Super-Duper") Arguments k type number seed kinds order lattice simple or triple lattice number of first plot seed method for to randomize Details kinds <.design. "Marsaglia-Multicarry". seed = 0.lsd.nk Examples library(agricolae) design. 2nd ed. design.c("Wichmann-Hill".

fact. design. The seed is by set. "user-supplied". New.lsd(trt.rcbd. "Random" uses the methods of number generation in R.crd. York. 1969 See Also design. seed = 0. kinds = "Super-Duper") Arguments trt Treatments number number of first plot seed seed kinds method for to randomize Details kinds <.ab.42 design.seed(seed. "Knuth-TAOCP-2002". "Mersenne-Twister". "Super-Duper".nk . "default" ) Value trt Vector names of treatments number Numeric seed Numeric Author(s) Felipe de Mendiburu References Introduction to Experimental Statistics. INC. Usage design. Ching Chun Li. "KnuthTAOCP". "Marsaglia-Multicarry". number = 1.c("Wichmann-Hill". McGraw-Hill Book Company.lsd Description It generates Latin Square Design. random.lsd Latin Square Design design. kinds).

"lsd. # write.rcbd(trt."maria bonita".rcbd Randomized Complete Block Design Description It generates Randomized Complete Block Design.design. r.4) print(t(plots)) print(t(trt)) # Write on hard disk. sep="\t") # file. numeric Numeric Numeric ."tomasa") lsd <-design. "Knuth-TAOCP-2002".seed(seed. "Super-Duper".txt") design. Usage design.numeric(lsd[.txt". number = 1. "user-supplied". "Random" uses the methods of number generation in R.names=FALSE.lsd(varieties.rcbd 43 Examples library(agricolae) varieties<-c("perricholi". The seed is by set. "KnuthTAOCP".seed=23) lsd # print field book. row. "Marsaglia-Multicarry". kinds = "Super-Duper") Arguments trt r number seed kinds Treatments Replications or blocks number of first plot seed method for to randomize Details kinds <. kinds). seed = 0. "default" ) Value trt r number seed vector. name of the treatments vector.c("Wichmann-Hill".4) dim(trt) <-c(4.4]) dim(plots)<-c(4.number=1001. plots <-as.1]) trt <-as.show("lsd.table(lsd. "Mersenne-Twister".character(lsd[."yungay".

design=c("rcbd".44 design. random.names=FALSE. row. York. seed = 0.split Split Plot Design Description It generates split plot design.show("rcbd."Super-Duper") # seed = 45 rcbd # field book plots <-as. Ching Chun Li."crd".5."lsd")."C".nk Examples library(agricolae) # 4 treatments and 5 blocks trt<-c("A".table(rcbd. INC. Usage design. McGraw-Hill Book Company. The seed is by set.split(trt1."rcbd.rcbd(trt."B". "Random" uses the methods of number generation in R.5) dim(trt) <-c(4.numeric(rcbd[.45. trt2. design.ab.txt") design.lsd. New."D") rcbd <-design.number=101. kinds).split Author(s) Felipe de Mendiburu References Introduction to Experimental Statistics.crd. fact.txt".r=NULL.seed(seed.character(rcbd[.3]) dim(plots)<-c(4. Arguments trt1 Treatments in Plots trt2 Treatments in Subplots r Replications or blocks design Experimental design .1]) trt <-as. sep="\t") # file.5) print(t(plots)) print(t(trt)) # write in hard disk # write. 1964 See Also design.number = 1.

"default" ) Value trt1 vector.kinds ="Super-Duper") # seed = 45 book # field book # write in hard disk # write.number=101. Kwanchai A.design."C". "Mersenne-Twister". fact. John Wiley & Sons. Gomez. "user-supplied". Arturo A. "KnuthTAOCP". 1984 See Also design. numeric design character number Numeric seed Numeric Author(s) Felipe de Mendiburu References Statistical Procedures for Agricultural Research.show("book.3) book <-design.table(book.c("Wichmann-Hill".t2. random. new York.txt") .2. design.split(t1. Gomez. row.r=3.txt"."book. "Knuth-TAOCP-2002". sep="\t") # file."D") t2<-c(1. "Marsaglia-Multicarry". numeric r vector. "Super-Duper".nk Examples library(agricolae) # 4 treatments and 5 blocks in split-plot t1<-c("A".lsd.split 45 number number of first plot seed seed kinds method for to randomize Details kinds <.crd.ab.seed=45.names=FALSE."B". numeric trt2 vector.

crd. kinds). seed = 0.c("Wichmann-Hill".seed(seed. Kwanchai A. "default" ) Value trt1 vector. kinds = "Super-Duper") Arguments trt1 Row treatments trt2 column treatments r Replications number number of first plot seed seed kinds method for to randomize Details kinds <.strip(trt1. trt2. design. "Mersenne-Twister". Gomez. Gomez.46 design. The seed is by set. "Super-Duper".nk . number = 1. "Marsaglia-Multicarry". name of the column treatments r numeric number Numeric seed Numeric Author(s) Felipe de Mendiburu References Statistical Procedures for Agricultural Research.ab. Arturo A.strip Description It generates strip plot design. new York. fact. "Random" uses the methods of number generation in R. 1984 See Also design.r. "user-supplied".lsd. "KnuthTAOCP".strip Strip Plot Design design. name of the row treatments trt2 vector. Usage design. random. "Knuth-TAOCP-2002". John Wiley & Sons.

"B". row.strip(t1. CIP .table(book.2. References International Potato Center. Examples library(agricolae) data(disease) str(disease) .show("book.names=FALSE. Usage data(disease) Format A data frame with 21 observations on the following 7 variables."C".seed=45.Lima Peru.r.t2.3) r<-3 book <-design.kinds ="Super-Duper") # seed = 45 book # field book # write in hard disk # write. plots a numeric vector rep a numeric vector trt a factor with levels T0 T1 T2 T3 T4 T5 T6 E2 a numeric vector E5 a numeric vector E7 a numeric vector yield a numeric vector Source Experimental data.txt") disease Data evaluation of the disease overtime Description Three evaluations over time and the potato yield when applying several treatments.disease 47 Examples library(agricolae) # 4 treatments and 5 blocks in split-plot t1<-c("A"."D") t2<-c(1."book. number=101. sep="\t") # file.txt".

DFerror.05. but only because it doesn’t control the error rate properly. main = NULL) Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degree free MSerror Mean Square Error alpha Significant level group TRUE or FALSE main Title Details It is necessary first makes a analysis of variance. The Experimentwise Error Rate at: 1-(1-alpha)^(a-1).test Description This test is adapted from the Newman-Keuls method. where "a" is the number of means and is the Per-Comparison Error Rate. The level by alpha default is 0.test Duncan’s new multiple range test duncan. alpha = 0. It has more power than the other post tests. Duncan’s test does not control family wise error rate at the specified alpha level.48 duncan. Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric alpha Numeric group Logic main Text Author(s) Felipe de Mendiburu . trt. Duncan’s procedure is only very slightly more conservative than LSD.05. group=TRUE. MSerror.test(y. Usage duncan.

duncan. Dealt with different virus") duncan. alpha = 0. HSD. Third Edition 1997 See Also LSD.test(judge. evaluation.05.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus. by default."virus".05. trt.test(model. group=TRUE) detach(sweetpotato) Durbin test and multiple comparison of treatments durbin.residual(model) MSerror<-deviance(model)/df comparison <. group =TRUE.test() attach(sweetpotato) df<-df. waller.test .test(model. Usage durbin.duncan.df. main="Yield of sweetpotato.test . It forms groups according to the demanded ones for level of significance (alpha).durbin.virus.test 49 References Principles and procedures of statistics a biometrical approach Steel & Torry & Dickey.test(yield."virus". main = NULL) Arguments judge Identification of the judge in the evaluation trt Treatments evaluation variable alpha level of significant group TRUE or FALSE main Title .01) # version old duncan.alpha=0.MSerror.test.data=sweetpotato) comparison <.test Description A multiple comparison of the Durbin test for the balanced incomplete blocks for sensorial or categorical evaluation. SNK. 0.

3.0.2.0. Myles Hollander.4. 0.3.test(person.0.2.602.0."G".20). Bliss.2) comparison<-durbin.931) comparison<-durbin.2. numeric evaluation Vector. pag 311 # Source: W.50 durbin. Conover. Conover.465.1.7.396.4."D"."A".536. friedman.0.group=TRUE.0.6.J. 1942 day<-gl(7.1.330."C".test Examples library(agricolae) # Example 1."F".2. 1999 See Also kruskal.0.3.5.1."B". numeric.2.6.preference. default is 0.group=TRUE.142. W."E".xlim=c(0.3.1.0.309.3.1.875.0.1. numeric alpha Vector.horiz=TRUE."C"."D".variety.3.0.609."C".343.0."E".3) variety<-c(1."F") toxic<-c(0. Moore and C.3.0. Myles Hollander and Douglas A.409.BIB."D".6.423.634."A".test Value juege Vector.1.0.5. pag 391 person<-gl(7."E". numeric trt Vector.987."F".989.7.4.group(comparison.0."B".toxic.1.652.3) chemical<-c("A".3.3. 1999 Nonparametric Statistical Methods.density=4) #endgraph # Example 2.test(day. main="Logarithm of Toxic Dosages") .0."G".2.2.426. Wofe. "B".873.0.3.05 group Logic main text Author(s) Felipe de Mendiburu References Practical Nonparametrics Statistics.7) preference<-c(2. main="Seven varieties of ice cream manufacturer") #startgraph bar.chemical.1.325."G".2.0.1.0.5.I.417.

design.crd. kinds). "KnuthTAOCP".nk 51 Factorial design k-factors with n-levels in blocks fact. Usage fact. design. "Super-Duper". "Random" uses the methods of number generation in R.rcbd.fact. design. factors. blocks. "Knuth-TAOCP-2002".nk Description It generates a table randomized for the design of blocks in factorial nk . kinds = "Super-Duper") Arguments level Number of levels factors Number of factor blocks Number of block seed Seed to randomize kinds method for to randomize Details kinds <. "Mersenne-Twister". "user-supplied".c("Wichmann-Hill".lsd. "Marsaglia-Multicarry". The seed is by set. random.nk(level. "default" ) Value level Numeric factors Numeric blocks Numeric seed Numeric kinds details Author(s) Felipe de Mendiburu See Also design.ab. seed = 0.ab .seed(seed.

factors<-3 .evaluation.2. blocks <.2.5.seed=100.kinds="Mersenne-Twister") # factorial 4x4 = 4^2 in 3 blocks fact.4 fact.nk(level.05.nk(3.4) # factorial two factors in 3 levels for 4 blocks fact.alpha=0. A first result is obtained by friedman.nk(2.52 friedman Examples library(agricolae) level<.trt.3) Friedman test and multiple comparison of treatments friedman Description The data consist of b mutually independent k-variate random variables called b blocks.blocks) # factorial 2^5 in 4 blocks fact.factors.main=NULL) Arguments judge Identification of the judge in the evaluation trt Treatment evaluation Variable alpha Significant test group TRUE or FALSE main Title Value judge Vector trt Vector evaluation Vector alpha Numeric group Logic main Text .nk(4.2.4.test of R.group=TRUE. It makes the multiple comparison of the Friedman test with or without ties. Usage friedman(judge. The random variable is in a block and is associated with treatment.

ylim=c(0.border="red". P2O5 and K2O + 2 t/ha of gallinaza. Tecnnologies: 20-40-20 kg/ha. main="Data of the book of Conover") #startgraph bar.45)) #endgraph frijol Data of frijol Description Data of frijol under 4 technologies for the homogeneity of regression study.frijol 53 Author(s) Felipe de Mendiburu References Practical Nonparametrics Statistics. 40-80-40 kg/ha. N.group(comparison. N. group=TRUE. W. N. P2O5 and K2O + 4 t/ha of gallinaza.trt. technology a factor with levels a b c d production a numeric vector index a numeric vector References Oriente antioqueno (1972) (ICA. 40-80-40 kg/ha.. P2O5 and K2O + 2 t/ha of gallinaza.) Colombia.J. N.alpha=0.col="blue".test Examples library(agricolae) data(grass) attach(grass) comparison<-friedman(judge.density=3. evaluation.05. Usage data(frijol) Format A data frame with 84 observations on the following 3 variables. Yield of Frijol in kg/ha in clean and dry grain. . durbin. 1999 See Also kruskal. 60-120-60 kg/ha. P2O5 and K2O + 2 t/ha of gallinaza.Orlando Martinez W. Conover.

Usage data(genxenv) Format A data frame with 250 observations on the following 3 variables. CIP .Lima Peru. Examples library(agricolae) data(genxenv) str(genxenv) .Lima Peru. References International Potato Center. CIP . ENV a numeric vector GEN a numeric vector YLD a numeric vector Source International Potato Center.54 genxenv Examples library(agricolae) data(frijol) str(frijol) genxenv Data of potato yield in a different environment Description 50 genotypes and 5 environments.

the graphic of frequency is obtained and it is possible to superpose the normal distribution. References International Potato Center. polygon of frequency. By means of this function.freq(x.Lima Peru. Usage data(Glycoalkaloids) Format A data frame with 25 observations on the following 2 variables. Examples library(agricolae) data(Glycoalkaloids) str(Glycoalkaloids) graph.frequency=1.xlab . HPLC a numeric vector spectrophotometer a numeric vector Source International Potato Center. nclass=NULL.counts=NULL. breaks="sturges". CIP .Glycoalkaloids Glycoalkaloids 55 Data Glycoalkaloids Description A measurement of the Glycoalkaloids by two methods: HPLC and spectrophotometer. Usage graph.freq Histogram Description In many situations it has intervals of class defined with its respective frequencies. plot=TRUE. Ojiva and to construct the table of complete frequency. CIP .Lima Peru.

xlab="YIELD" axis(1.col="bisque". other parameters of graphic Value x Numeric or list plot TRUE or FALSE counts Numeric frequency Numeric Author(s) Felipe de Mendiburu See Also polygon.7.freq.freq.freq() counts frequency and x is class intervals breaks a vector giving the breakpoints between histogram cells frequency 1=counts.col="yellow") axis(1. ylab="frequency".graph.counts=frequencies.freq Arguments x a vector of values.20. 2=relative. only frecuency limits <-seq(10.seq(0. frequency=2.1) h<.. sturges.freq(h.round(yield.axes=FALSE. 3=density plot logic nclass number of classes xlab x labels ylab y labels .5) frequencies <-c(2.border="red".freq.freq.3.4.genxenv$ENV==2) yield <. normal.1)) # summary.axes=FALSE.xlab="Classes") polygon.56 graph.col="red") .freq(limits. intervals.graph.freq.0. join. a object hist(). density=8.0. stat...freq.freq. table. ojiva. col="blue".freq Examples library(agricolae) data(genxenv) yield <.8.6.subset(genxenv$YLD.h$breaks) axis(2.4) #startgraph h<-graph.h$breaks) axis(2. h1 <.freq(yield. frequency=1.freq(h.40. graph.1)) # To reproduce histogram.seq(0.

ylab="Accumulated relative frequency") # only frequency polygon h<-graph.seq(0.grass 57 title( main="Histogram and polygon of frequency".xlab="Variable". Usage data(grass) Format A data frame with 48 observations on the following 3 variables.col="blue") grid(col="brown") #endgraph # Draw curve for Histogram h<.0.col="red". lwd = 2) title("Draw curve for Histogram") grass Data for Friedman test Description Twelve homeowners are selected randomly to participate in an experiment with a plant nursery. Each homeowner is asked to select four fairly identical areas in his yard and to plant four different types of grasses. ylab="Frecuency") polygon.col=NULL.type="b".border=FALSE. each receiving care of approximately the same degree of skill because the four plots are presumably cared for by the same homeowern.freq(h).xlab=" ".ylab="") title( main="Polygon of frequency".counts=frequencies.freq(limits. col = "red".h$breaks) axis(2.18. . xlab="Variable".col="yellow") axis(1.axes=FALSE.las=2) lines(density(yield). ylab="Frequency") #endgraph # Statistics measures<-stat.freq(yield.graph.freq(h.freq(h) print(measures) # frequency table full round(table. judge a numeric vector trt a factor with levels t1 t2 t3 t4 evaluation a numeric vector Details Each of the 12 blocks consists of four fairly identical plots of land.03).freq(h. one in each area. ylab="f(yield)".2) #startgraph # Ojiva ojiva. frequency=3.0.

References Practical Nonparametrics Statistics. AGROENFOQUE Lima-Peru (2004) . Source International Potato Center(CIP).Produccion de semila de papa por hidroponia tecnica de flujo continuo de una pelicula de solucion nutritiva (nft) Carlos Chuquillanqui(CIP). planting date(April 24. the unit is in grams and the number of tubers in units. three sets of data in potato varieties with different methods: hydroponics. Jorge Tenorio(CIP) and L. greenhouse2(48 obs. unit is grams. Conover. F. 5 var).J. Imma. 2004) and greenhouse3(480 obs. Salazar(Agdia Inc). 5 var).58 greenhouse Source Book: Practical Nonparametrics Statistics. Data Kindly provided by Carlos Chuquillanqui. Jan/Feb 2006 by Farran. 5 var). Angel M . W. Pots and Plant beds. Mingo-Castel. Aeroponic. 1999 Examples data(grass) str(grass) greenhouse Data in greenhouse Description Potato minituber production in greenhouse. 2004) and harvest date(July 16. Lima-Peru. Yields of 10 plants by experimental unit(grams). Tubers by plants. three tables: greenhouse1(480 obs. References . Usage data(greenhouse) Details greenhouse is list.Potato Minituber Production Using Aeroponics: Effect of Plant Density and Harvesting Intervals American Journal of Potato Research. yield for plant. pag 372.

y) generates an information matrix by a process of interpolation for nonequidistant points.greenhouse$greenhouse2 greenhouse3 <. m. those that will be obtained by interpolation. Value x Numeric y Numeric z Numeric m Numeric n Numeric Author(s) Felipe de Mendiburu . y. It uses function interpp of library akima.grid3p 59 Examples library(agricolae) data(greenhouse) greenhouse1 <. n) Arguments x Vector independent y Vector independent z Vector dependent m number of rows of the new matrix n number of columns of the new matrix Details The function fxyz obtains a new data set.greenhouse$greenhouse1 greenhouse2 <. A new vector "x" of "m" elements and a new vector "y" of "n" elements and the matrix "z" of mxn elements.greenhouse$greenhouse3 Interpolation for nonequidistant points in matrix grid3p Description Z=f(x. Usage grid3p(x. z.

rep(b[i]. col = "blue".xlab="clay".add=TRUE.32.9.y.xo=seq(4.clay y<-clay$days z<-clay$ralstonia model<.cex=0.7.z) # it completes and it finds the average of points with equal coordinate.m. ylab = "day".z2.side=3.length=100).numeric(zo) m<-40 n<-20 # It generates a new matrix mxn with but points by interpolation.side=3.colors(8)) contour(zz$x.zz$y.79.interp(x. z2<-grid3p(x. .frame=FALSE.lm(z ~ x + y) zo<-wxyz(model.60 growth See Also wxyz Examples library(akima) library(agricolae) data(clay) x<-clay$per.length=100).y.z.y.z.numeric(dimnames(z2)[[1]]) y2<-as.numeric(dimnames(z2)[[2]]) res<-contour(x2.ylab="days") mtext("Ralstonia solanacearum population". cex=0.7. col=topo. col="blue".frame=FALSE) mtext("Ralstonia solanacearum population\n".zz$z.zz$z.x.length(a))) y<-as.clay y<-clay$days z<-clay$ralstonia zz <.font=4) #==================== # Using the function of interpolacion of irregular points.length(b))) y<-NULL for(i in 1:length(b)) y<-c(y.y2.xlab = "clay".zz$y. see interp() de "akima" data(clay) x<-clay$per. b<-colnames(zo) a<-rownames(zo) x<-as.font=4) #endgraph growth Data growth of trees Description Data growth of pijuayo trees in several localities.numeric(rep(a.yo=seq(2.numeric(y) z<-as.duplicate="mean") #startgraph image(zz$x.cex=0.9.n) # plot x2<-as. cex=0.

Peru) References ICRAF lima Peru. place a factor with levels L1 L2 slime a numeric vector height a numeric vector Source Experimental data (Pucallpa . Usage data(haynes) Format A data frame with 16 observations on the following 9 variables. Haynes. Examples library(agricolae) data(growth) str(growth) haynes Data of yield for nonparametrical stability analysis Description Published data. clone a factor with levels A84118-3 AO80432-1 AO84275-3 AWN86514-2 B0692-4 B0718-3 B0749-2F B0767-2 Bertita Bzura C0083008-1 Elba Greta Krantz Libertas Stobrawa FL a numeric vector MI a numeric vector ME a numeric vector MN a numeric vector .haynes 61 Usage data(growth) Format A data frame with 30 observations on the following 3 variables.

62 Hco2006 ND a numeric vector NY a numeric vector PA a numeric vector WI a numeric vector References Haynes K G. Christ B J. Phenotypic stability of resistance to late blight in potato clones evaluated at eight sites in the United States American Journal Potato Research 75. 1998. pag 211-217. kg/8m2. replications plant a numeric vector. number plant wilt_percent a numeric vector. Examples library(agricolae) data(haynes) str(haynes) Hco2006 Data amendment Huanuco 2006 Description Incidents and performance of healthy tubers and rotten potato field infested with naturally Ralstonia solanacearum Race 3/Bv 2A. Usage data(Hco2006) Format The format is: List of 2 amendment a factor crop a factor block a numeric vector. 2006. 20 plants rot a numeric vector. 20 plants Details Application of inorganic amendment and crop rotation to control bacterial wilt of the potato (MBP). Douches D S. . Lambert D H. Fry W and Stevenson W. Weingartner D P. kg/8m2. after application of inorganic amendments and a rotation crop in Huanuco Peru. wilt percentage at 60 days health a numeric vector. Backlund J E.

method="spearman") hcut Cut tree of consensus Description It shows dendrogram of a consensus of a tree generated by hclust.text=1.h.col.text color of number consensus cex.5].group.hcut 63 Source Experimental field.text="blue"..3])).model(model) yield<-yield[order(paste(yield[.tapply.yield[. 2006.4].aov(wilt_percent ~ block + crop.frame with group memberships and consensus tree.text size of number consensus .. CIP . Examples library(agricolae) data(Hco2006) str(Hco2006) wilt<-Hco2006$wilt yield<-Hco2006$yield means <. Data Kindly provided by Pedro Aley.wilt[.yield[. .. Usage hcut(consensus. Other parameters of the function plot() in cut() Value hcut returns a data. means) anova(model) cv.Lima Peru.1:3]. References International Potato Center.stat(wilt[.na.rm=TRUE)) names(means)[4]<-"wilt_percent" model <.cex.function(x) mean(x.yield[.1]...] correlation(means[.2]. group an integer scalar with the desired number of group col.4].) Arguments consensus object consensus h numeric scalar or vector with heights where the tree should be cut.

h=0. Usage data(heterosis) Format A data frame with 216 observations on the following 11 variables.h=0. de Mendiburu See Also hclust.64 heterosis Author(s) F. Heterosis Description Determination of heterosis.type="t".1. col=2:1). hgroups Examples library(agricolae) data(pamCIP) # only code rownames(pamCIP)<-substr(rownames(pamCIP).4.group=5.cex.main="group 8" .main="Group 5") # hcut(output.nboot=100) hcut(output.text=1) heterosis Data of potato. Place 1: La Molina.group=8. 2=Huancayo Replication a numeric vector Treatment a numeric vector Factor a factor with levels Control progenie progenitor testigo Female a factor with levels Achirana LT-8 MF-I MF-II Serrana TPS-2 TPS-25 TPS-7 Male a factor with levels TPS-13 TPS-67 TS-15 v1 Yield (Kg/plant) v2 Reducing sugars (scale):1 low and 5=High v3 Tuber dry matter (percentage) v4 Tuber number/plant v5 Average tuber weight (g) .text="blue".col.4. reducing sugars and agronomic characteristics in TPS families.edgePar = list(lty=1:2. consensus.6) # groups of clusters output<-consensus(pamCIP. general combining ability (GCA) and specific combining ability in tuber dry matter.

HSD. The experimental material consisted of 24 families half brother in the form of tubers derived from TPS. Source International Potato Center(CIP). Poster: Congreso de la Sociedad Peruana de Genetica . Universidad Nacional Agraria La molina-Lima Peru. v3) detach(site2) attach(site3) model3<-lineXtester(Replication. Variables evaluated were Yield. La Molina-PERU to 240 masl. Female. Dry matter and content and reducing sugars.test(v1.heterosis[. v4) detach(site3) hgroups groups of hclust .hgroups 65 Details The study was conducted in 3 environments. azucares reductores y caracteres agronomicos en familias de tuberculos provenientes de semilla sexual de papa.. Magister Scientiae Rodolfo Valdivia Lorente.1]==2) site3<-subset(heterosis. v1) DFe <. Female.residual(model1) CMe <. Male. References Tesis "Heterosis.DFe.CMe) test2 <.deviance(model1)/DFe test1 <.CMe) detach(site1) attach(site2) model22<-lineXtester(Replication. during summer.df. Design used was randomized complete block with three repetitions. Female.heterosis[. and in Huancayo-PERU 3180 masl.test(v1. Male. The control variety was Desiree. obtained crossing between 8 female and 3 male parents. Mejoramiento genetico de plantas.HSD. 2008.1]==1) site2<-subset(heterosis. Lima-Peru. Tubers number. Data Kindly provided by of Rolando Cabello. Escuela de Post Grado. Factor. Examples library(agricolae) data(heterosis) str(heterosis) site1<-subset(heterosis.DFe. during autumn-winter and spring. 2004". Treatment. habilidad combinatoria general y especifica para materia seca. The analysis was conducted line x tester. Male.Peru.1]==3) attach(site1) model1<-lineXtester(Replication.heterosis[. The experimental unit was 30 plants in two rows at a distance of 30cm between plants and 90 cm between rows.

1. group=TRUE.05. DFerror. The level by alpha default is 0. de Mendiburu See Also hclust. alpha = 0. method="complete") # groups of clusters hgroups(clusters$merge) HSD.test Description Returns a vector with group memberships.05.6) distance <.66 HSD. This function is used by the function consensus of clusters. MSerror.method="binary") clusters<.hclust( distance. consensus Examples library(agricolae) data(pamCIP) # only code rownames(pamCIP)<-substr(rownames(pamCIP).test Multiple comparisons: Tukey Description It makes multiple comparisons of treatments by means of Tukey. Usage HSD.test(y. Usage hgroups(hmerge) Arguments hmerge The object is components of the hclust Value data object merge of hcluster’ Author(s) F.dist(pamCIP. trt. hcut. main = NULL) .

test . SNK. data=sweetpotato) comparison <. main="Yield of sweetpotato\nDealt with different virus") #stargraph bar.border="blue") #endgraph .test 67 Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degree free MSerror Mean Square Error alpha Significant level group TRUE or FALSE main Title Details It is necessary first makes a analysis of variance.group(comparison.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus.HSD.test(model.test.45).ylim=c(0.test .density=4. waller. Third Edition 1997 See Also LSD."virus". group=TRUE. Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric alpha Numeric group Logic main Text Author(s) Felipe de Mendiburu References Principles and procedures of statistics a biometrical approach Steel & Torry & Dickey. duncan.HSD.

main="Yield of sweetpotato. Dealt with different virus") detach(sweetpotato) huasahuasi Data: Rainfall thresholds as support for timing fungicide applications in the control of potato late blight in Peru Description Timing fungicide sprays based on accumulated rainfall thresholds can be a successful component of integrated management packages that include cultivars with moderate or high levels of resistance to late blight.8 m of length. The simplicity of measuring accumulated rainfall means that the technology can potentially be used by resource-poor farmers in developing countries. The experiment had 3 repetitions. Without application: No fungicide application will be made. Kgr. The . Usage data(huasahuasi) Format The format is: List of 2 ( AUDPC. In each furrow was installed 5 plants. Kgr. with distance between furrows from 0.test(model.test(yield. YIELD ) block a factor with levels I II III trt a factor with levels 40mm 7-days Non-application clon a factor with levels C386209.90 m and between plants of 0.residual(model) MSerror<-deviance(model)/df HSD. From the beginning of the experiment were fulfilled the following treatments Thresholds 40 mm: Apply the fungicide when 40 precipitation mm accumulates.68 huasahuasi HSD. group=TRUE. 51 days d100 a numeric vector. Kgr./plot y2da a numeric vector.test() attach(sweetpotato) df<-df.10 C387164.4 Cruza148 Musuq Yungay y1da a numeric vector.MSerror. 100 days Details The experimental unit was formed by 4 furrows of 1."virus". The minimum interval between applications will be of 7 days. Schedule 7 days: The applications should be carried out every 7 days calendar. 44 days d51 a numeric vector./plot d44 a numeric vector.30 m.virus./plot y3ra a numeric vector. group=FALSE) # Old version HSD.df.

Parker" "McIntosh" "Shannon" .bio(data.bio Biodiversity Index Description Scientists use a formula called the biodiversity index to describe the amount of species diversity in a given area.Div" "Berger.Parker Arguments data number of specimens method Describe method bio-diversity level Significant level nboot size bootstrap console output console TRUE Details method bio-diversity. "Simpson.Lima Peru. CIP .Div". "Simpson.bio 69 evaluation of the severity of the late blight in each treatment started to emergency 80 percentage and then evaluations were made every 7 days until being observed a physiological maturation of the crop. Usage index. References International Potato Center. 2003. Source Experimental field. "Berger.Dom". method = c("Margalef".Dom" "Simpson. Data Kindly provided by Wilmer Perez. "Margalef" "Simpson. Examples library(agricolae) data(huasahuasi) names(huasahuasi) str(huasahuasi$AUDPC) str(huasahuasi$YIELD) index.index.

nboot=200) output2 <.nboot=200) rbind(output1.level=95. The index gives a single value as a quantitative measure of soil heterogeneity in an area.level=95..smith Value data method level nboot vector method bio-diversity value 0-100 percentage size 100.... and b is the Smith’ index of soil heterogeneity.smith(data.method="Shannon".. 500.bio(specimens. Usage index.6] output1 <.E.bio(specimens. Smith’s Index of Soil Heterogeneity Description Smith’s index of soil heterogeneity is used primarily to derive optimum plot size. Graph CV for plot size and shape.smith Uniformity soil.paracsho[1:10. R. Tibshirani.. A. Vx is the variance per unit area for plot size of x basic unit. .Div".index. Chapman and Hall/CRC Examples library(agricolae) data(paracsho) # date 22-06-05 and treatment CON = application with insecticide specimens <. Princeton University Press Efron.index.method="Simpson. (1993) An Introduction to the Boostrap.. B. output2) index. (1988) Ecological diversity and its measurement..70 index. dataframe or matrix Parameters of the plot() Details Vx=V(x)/x b V(x) is the between-plot variance.) Arguments data . . Author(s) Felipe de Mendiburu References Magurran.

freq 71 Value data Numeric . Gomez and Arturo A. main="Relationship between CV per unit area and plot size"..Vx. col="blue". Gomez. Author(s) Felipe de Mendiburu References Statistical Procedures for Agriculture Research. type="l".freq Class intervals Description List class intervals. Kwanchai A. 1976.coef(model) x<-1:max(uniformity$Size) Vx<.coeff[1]+coeff[2]*log(x) #startgraph plot(x.intervals.smith(rice.lm(Vx ~ I(log(Size)).data.uniformity) coeff <.uniformity$Vx) #endgraph intervals.col="red") #endgraph uniformity <.frame(table$uniformity) uniformity # regression variance per unit area an plot size. Second Edition. model <. Usage intervals..freq(breaks) Arguments breaks class intervals . USA Examples library(agricolae) data(rice) #startgraph table<-index. main="Relationship between variance per unit area and plot size") points(uniformity$Size.

freq(h2$breaks) # example 2 x<-seq(10.freq Description In many situations it is required to join classes because of the low frequency in the intervals. graph.freq(x) Join class for histogram join.6. join.72 join.40. stat. In this process.7. join) Arguments breaks Class intervals join vector Value breaks vector numeric join numeric .freq. normal. sturges. table.freq. Usage join.freq.freq(breaks.8.freq. ojiva.freq.4) intervals.plot=FALSE) intervals.5) y<-c(2.freq. it is required to join the intervals and ad the frequencies of them.freq Value breaks vector numeric Author(s) Felipe de Mendiburu See Also polygon.3.freq Examples library(agricolae) # example 1 data(growth) attach(growth) h2<-hist(height.freq.

breaks.kendall 73 Author(s) Felipe de Mendiburu See Also polygon.0.freq. inter<-join. Usage kendall(data1. ojiva.9:10) # new table h2<-hist(size.freq(breaks. Compute exact p-value with ties.12). stat. data2) Arguments data1 vector data2 vector Value data1 Numeric data2 Numeric .plot=FALSE) table.xlim=c(0.freq. normal. table.freq.freq.breaks=inter.class$breaks # list classes intervals.col="bisque") table. sturges.freq(size) # information of the classes class breaks <.freq Examples library(agricolae) data(natives) attach(natives) class<-sturges.freq.freq(h1) # Join classes 9 and 10 with little frequency. graph.freq. intervals.freq(breaks) # Table frecuency h1<-hist(size.freq.freq(h2) Correlation of Kendall kendall Description Correlation of Kendall two set.

Second Edition.1.3.2) y <-c(1.1. Pag 634 See Also correlation Examples library(agricolae) x <-c(1.2.3.1.1.2) kendall(x.3.2.1.1.4. The parameters by default are alpha = 0.2.3.1.2.4.1. group=TRUE.2.05.2. main = NULL) Arguments y response trt treatment alpha level signification group TRUE or FALSE main Title Value y vector numeric trt vector alphanumeric alpha level significant group Logic main Title .y) Kruskal Wallis test and multiple comparison of treatments.4.3.4.3. trt.2.1. kruskal Description It makes the multiple comparison with Kruskal-Wallis.2. Usage kruskal(y.1.1.74 kruskal Author(s) Felipe de Mendiburu References Numerical Recipes in C.2.3. alpha = 0.1.1.1.05.1.2.

5. The result is similar to SAS. W. durbin.3.NA.4.7) kurtosis(x) # value is -0.method. Conover.6.5.2.4.group=TRUE.4.1517996 .J.kurtosis 75 Author(s) Felipe de Mendiburu References Practical Nonparametrics Statistics.test Examples library(agricolae) data(corn) attach(corn) str(corn) comparison<-kruskal(observation. Usage kurtosis(x) Arguments x a numeric vector Value x The kurtosis of x See Also skewness Examples library(agricolae) x<-c(3. 1999 See Also friedman. main="corn") detach(corn) kurtosis Finding the Kurtosis coefficient Description It obtains the value of the kurtosis for a normally distributed variable.

"cd") lastC(x) # "a" "b" "b" "c" "d" lineXtester Line x Tester Analysis Description It makes the Line x Tester Genetic Analysis. Usage lastC(x) Arguments x letters Value x character Author(s) Felipe de Mendiburu See Also order.76 lineXtester Setting the last character of a chain lastC Description A special function for the group of treatments in the multiple comparison tests. It also estimates the general and specific combinatory ability effects and the line and tester genetic contribution. Use order. Usage lineXtester(replications."c". lines."ab"."b".group.group Examples library(agricolae) x<-c("a". testers. y) .

Componentes geneticos. to total variance. v2) detach(site1) # example 2 data(LxT) str(LxT) attach(LxT) output2<-lineXtester(replication. Genetic Components. response Details output: Standard Errors for combining ability effects. Proportional contribution of lines. ANOVA for line X tester analysis. yield) detach(LxT) . Female. ANOVA for line X tester analysis including parents. Standard Errors for Combining Ability Effects. Testers Effects. tester.heterosis[. Variancias. numeric vector. numeric Author(s) Felipe de Mendiburu References Biometrical Methods in Quantitative Genetic Analysis. Chaudhary. Lines Effects. line. Singh.1]==1) attach(site1) output1<-lineXtester(Replication. GCA Effects. numeric vector. Male. testers and their interactions. ANOVA with parents and crosses. numeric vector. 1979 See Also AMMI Examples # see structure line by testers library(agricolae) # example 1 data(heterosis) site1<-subset(heterosis. Value replications lines testers y vector. Contribucion proporcional.lineXtester 77 Arguments replications lines testers y Replications Lines Testers Variable.

"BH".05. MSerror. group=TRUE.adj=c("none". "BH".78 LSD.05. alpha = 0. p. main = NULL) Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degrees of freedom of the experimental error MSerror Means square error of the experimental alpha Level of risk for the test p.test Description Multiple comparisons of treatments by means of LSD and a grouping of treatments. DFerror.adj text.adj = "holm".adjust group Logic main Numeric Author(s) Felipe de Mendiburu .test(y.adjust() p-adj ="none" is t-student. "BY"."holm". "bonferroni". p-adj ="hommel" is not applied in this test. "hochberg". "fdr". The level by alpha default is 0. "bonferroni". trt. "Least significant difference" and Adjust Pvalues LSD. "fdr"). see p.test Multiple comparisons. "hochberg".adj Method for adjusting p values (see p. Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric alpha Numeric p. see p.adjust) group TRUE or FALSE main title of the study Details For equal or different repetition. "BY". p. Returns p-values adjusted using one of several methods Usage LSD.

err. pp178.test.virus.LSD.group.LxT 79 References Steel. main="Yield of sweetpotato\ndealt with different virus") detach(sweetpotato) #stargraph bar. bar.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus."virus".ylim=c(0.border="red") #endgraph LSD. See Also HSD. Torri.test() attach(sweetpotato) df<-df.density=4.adj="bonferroni".test(yield. p.duncan.. Dickey.p. data=sweetpotato) comparison <.MSerror.df.test.(1997) Principles and Procedures of Statistics A Biometrical Approach.test(model. waller.LSD. group=FALSE. Usage data(LxT) Format A data frame with 92 observations on the following 4 variables.group(comparison.J. bar. D.45). main="Yield of sweetpotato\ndealt with different virus") #stargraph bar."virus". replication a numeric vector line a numeric vector tester a numeric vector yield a numeric vector .adj="bonferroni".err(comparison.border="blue") #endgraph # Old version LSD.45). p. R. SNK.test .std=TRUE.adj="bon") LxT Data Line by tester Description Data frame with yield by line x tester.ylim=c(0.test(model.residual(model) MSerror<-deviance(model)/df comparison <.density=4.

Singh. marker1 a numeric vector marker2 a numeric vector marker3 a numeric vector marker4 a numeric vector marker5 a numeric vector marker6 a numeric vector marker7 a numeric vector marker8 a numeric vector marker9 a numeric vector marker10 a numeric vector marker11 a numeric vector marker12 a numeric vector marker13 a numeric vector marker14 a numeric vector marker15 a numeric vector marker16 a numeric vector marker17 a numeric vector marker18 a numeric vector marker19 a numeric vector marker20 a numeric vector marker21 a numeric vector marker22 a numeric vector .80 markers Source Biometrical Methods in Quantitative Genetic Analysis. Usage data(markers) Format A data frame with 23 observations on the following 27 variables. Chaudhary. 1979 markers Data of molecular markers Description A partial study on 27 molecular markers.

Usage data(melon) Format A data frame with 16 observations on the following 4 variables. Examples library(agricolae) data(markers) str(markers) melon Data of yield of melon in a Latin square experiment Description An irrigation system evaluation by exudation using four varieties of melon. Hibrido Mark (V2). References International Potato Center Lima-Peru. . Hibrido Topfligth (V3). The goal is to analyze the behavior of three hybrid melon varieties and one standard.melon 81 marker23 a numeric vector marker24 a numeric vector marker25 a numeric vector marker26 a numeric vector marker27 a numeric vector Source International Potato Center Lima-Peru. row a numeric vector col a numeric vector variety a factor with levels V1 V2 V3 V4 yield a numeric vector Details Varieties: Hibrido Mission (V1). under modality of sowing. SIMPLE ROW. Hibrido Hales Best Jumbo (V4).

Universidad Agraria la Molina Lima Peru. .. k. Examples library(agricolae) data(melon) str(melon) Random generation by Montecarlo montecarlo Description Random generation form data.) Arguments data vector or object(hist.82 montecarlo Source Tesis. graph. SIMPLE HILERA".freq) k number of simulations . only if data is vector Value Generates random.. use function density and parameters Usage montecarlo(data. References Universidad Nacional Agraria la molina. Other parameters of the function density... Alberto Angeles L. "Evaluacion del sistema de riego por exudacion utilizando cuatro variedades de melon. Author(s) Felipe de Mendiburu See Also density . bajo modalidad de siembra.

k=100) natives Data of native potato Description An evaluation of the number. Usage data(natives) Format A data frame with 876 observations on the following 4 variables. CIP . k=100) # other example breaks<-c(0. 300) counts<-c(10. 10. 30) h<-graph. 20.plot=FALSE) montecarlo(h. weight and size of 24 native potatoes varieties. variety a numeric vector number a numeric vector weight a numeric vector size a numeric vector Source International Potato Center. Examples library(agricolae) data(natives) str(natives) . 150.Lima Peru.freq(x=breaks. 250.natives 83 Examples library(agricolae) r<-rnorm(50. kernel="epanechnikov") # other example h<-hist(r. 200.plot=FALSE) montecarlo(h.2) montecarlo(r. k=100. 40.counts=counts.

84

nonadditivity

Nonadditivity model test

nonadditivity

Description
The resistance for the transformable nonadditivity, due to J. W. Tukey, is based on the detection of
a curvilinear relation between y-est(y) and est(y). A freedom degree for the transformable nonadditivity.
Usage
nonadditivity(y, factor1, factor2, df, MSerror)
Arguments
y

Answer of the experimental unit

factor1

Firts treatment applied to each experimental unit

factor2

Second treatment applied to each experimental unit

df

Degrees of freedom of the experimental error

MSerror

Means square error of the experimental

Details
Only two factor: Block and treatment or factor 1 and factor 2.
Value
y

Numeric

factor1

alfanumeric

factor2

alfanumeric

df

Numeric

MSerror

Numeric

Author(s)
Felipe de Mendiburu
References
1. Steel, R.; Torri,J; Dickey, D.(1997) Principles and Procedures of Statistics A Biometrical Approach
2. George E.P. Box; J. Stuart Hunter and William G. Hunter. Statistics for experimenters. Wile
Series in probability and statistics

normal.freq

85

Examples
library(agricolae)
data(potato )
potato[,1]<-as.factor(potato[,1])
model<-lm(cutting ~ date + variety,potato)
df<-df.residual(model)
MSerror<-deviance(model)/df
attach(potato)
analysis<-nonadditivity(cutting, date, variety, df, MSerror)
detach(potato)

Normal curve on the histogram

normal.freq

Description
A normal distribution graph elaborated from the histogram previously constructed. The average and
variance are obtained from the data grouped in the histogram.
Usage
normal.freq(histogram, frequency=1, ...)
Arguments
histogram

object constructed by the function hist

frequency

1=counts, 2=relative, 3=density

...

Other parameters of the function hist

Value
Histogram

object

frequency

numeric

probability

logic False or True

Author(s)
Felipe de Mendiburu
See Also
polygon.freq, table.freq, stat.freq, intervals.freq, sturges.freq, join.freq,
ojiva.freq, graph.freq

86

ojiva.freq

Examples
library(agricolae)
data(growth)
attach(growth)
#startgraph
h1<-hist(height,col="green",xlim=c(6,14))
normal.freq(h1,col="blue")
#endgraph
#startgraph
h2<-graph.freq(height,col="yellow",xlim=c(6,14),frequency=2)
normal.freq(h2,frequency=2)
#endgraph

Plotting the ojiva from a histogram

ojiva.freq

Description
It plots the cumulative relative frequencies with the intervals of classes defined in the histogram.
Usage
ojiva.freq(histogram, ...)
Arguments
histogram

object created by the function hist() or graph.freq()

...

Parameters of the hist()

Value
histogram

Object

...
Author(s)
Felipe de Mendiburu
See Also
polygon.freq, table.freq, stat.freq, intervals.freq, sturges.freq, join.freq,
graph.freq, normal.freq

Kruskal-Wallis. This function is used by the HSD. 0. means. std. xlab="Limit of class".order.freq(h. MSerror.frame=FALSE.col="red". Friedman or Durbin procedures. N.err standard error parameter Constante 1 (Sd).plot=FALSE) points<-ojiva. Tprob. parameter=1.type="l". Usage order.group(trt.group Ordering the treatments according to the multiple comparison Description This function allows us to compare the treatments averages or the adding of their ranges with the minimal significant difference which can vary from one comparison to another one.err. ylab="Accumulated relative frequency".5 (Sx) snk Constante = 1 (Student Newman Keuls) DFerror Degrees of freedom of the experimental error alpha Level of risk for the test sdtdif standar deviation of difference in BIB . main="ojiva") grid(col="black") #endgraph print(points) order. LSD.group 87 Examples library(agricolae) data(growth) attach(growth) #startgraph h<-graph. snk=0.freq(height. DFerror=NUL Arguments trt Treatments means Means of treatment N Replications MSerror Mean square error Tprob minimum value for the comparison std.

err Numeric parameter Constant snk Constant DFerror Numeric alpha Numeric sdtdif Numeric Author(s) Felipe de Mendiburu See Also order. 1. 3.err(groups.stat Examples library(agricolae) treatments <.2.1314 groups<-order.MSerror.t <.35.means. means. 2.80)) order.4.ylim=c(0.8 value.1) replications <.55. minimum) .3) MSerror <.c("A".4.5."D".replications.2.4. these can be compared with a minimal difference of meaning.group(treatments. 2.88 order. 1."E"."F") means<-c(20.stat Grouping the treatments averages in a comparison with a minimum value Description When there are treatments and their respective values.value."C".3.stat Value trt Factor means Numeric N Numeric MSerror Numeric Tprob value between 0 and 1 std.58) std. Usage order.std.72.err<-c(1.40.c(4."B".3.t.err) groups bar.stat(treatment.49.

2 groups<-order.3.c("A".5.stat(treatments.means.group Examples library(agricolae) treatments <. Rownames: code and genotype’s name. Details To study the molecular markers in Wild."D".9.pamCIP 89 Arguments treatment treatment means means of treatment minimum minimum value for the comparison Value trt Factor means Numeric minimum Numeric Author(s) Felipe de Mendiburu See Also order.minimum."E". ."F") means<-c(2.diff) pamCIP Data Potato Wild Description Potato Wild Usage data(pamCIP) Format A data frame with 43 observations on the following 107 variables.5) minimum.diff <."B"."C".7. column data: molecular markers.

Deparment Junin. . A biodiversity.specimens a numeric vector Details Country Peru. References International Potato Center Lima-Peru (CIP) Examples library(agricolae) data(pamCIP) str(pamCIP) paracsho Data of Paracsho biodiversity Description A locality in Peru. Usage data(paracsho) Format A data frame with 110 observations on the following 6 variables. province Tarma. locality Huasahuasi. date a factor with levels 15-12-05 17-11-05 18-10-05 20-09-05 22-06-05 23-0805 28-07-05 plot a factor with levels PARACSHO Treatment a factor with levels CON SIN Orden a factor with levels COLEOPTERA DIPTERA HEMIPTERA HYMENOPTERA LEPIDOPTERA NEUROPTERA NEUROPTERO NOCTUIDAE Family a factor with levels AGROMYZIDAE ANTHOCORIDAE ANTHOMYIIDAE ANTHOMYLIDAE BLEPHAROCERIDAE BRACONIDAE BROCONIDAE CALUPHORIDAE CECIDOMYIDAE CHENEUMONIDAE CHNEUMONIDAE CHRYOMELIDAE CICADELLIDAE CULICIDAE ERIOCPAMIDAE HEMEROBIIDAE ICHNEUMONIDAE LOUCHAPIDAE MIRIDAE MUSCIDAE MUSICADAE MUSLIDAE MYCETOPHILIDAE MYCETOPHILIIDAE NENPHALIDAE NOCLUIDAE NOCTERIDAE NOCTUIDAE PERALIDAE PIPUNCULIDAE PROCTOTRUPIDAE PSYLLIDAE PYRALIDAE SARCOPHAGIDAE SARCOPILAGIDAE SCATOPHAGIDAE SCATOPHOGIDAE SCIARIDAE SERSIDAE SYRPHIDAE TACHINIDAE TIPULIDAE Number.of.90 paracsho Source Laboratory data.

1979 .x Matrix of correlations of the independent variables corr. References International Potato Center. The function calculates the direct and indirect effects and uses the variables correlation or covariance. it is possible to represent the whole system of variables in a diagram form known as path-analysis.x Numeric corr. Value corr.analysis(corr.path. Chaudhary.y) Arguments corr.x. corr.y Numeric Author(s) Felipe de Mendiburu References Biometrical Methods in Quantitative Genetic Analysis. Examples library(agricolae) data(paracsho) str(paracsho) Path Analysis path. Usage path. Singh.analysis Description If the cause and effect relationship is well defined.y vector of dependent correlations with each one of the independent ones Details It is necessary first to calculate the correlations.analysis 91 Source Entomology dataset.

matrix(c(1."Y") path.test Description Analysis of variance PBIB and comparison mean adjusted.7) names<-c("X1". alpha=0.12.y) Analysis of the Partially Balanced Incomplete Block Design PBIB.5. Anderson.test See Also correlation Examples # Path analysis.c(2.0.6. method = c("lsd". Prentice Hall."tukey").05) Arguments block blocks trt Treatment replication Replication y Response k size block method Comparison treatments alpha Significant test . Usage PBIB.corr.y<."X2") dimnames(corr. Applied to resoluble designs: Lattices and alpha design.x)$correlation cor.trt.cor.replication.1).y<-correlation(y.y) # Example 2 # data of the progress of the disease related bacterial wilt to the ground # for the component CE Ca K2 Cu data(wilt) data(soil) x<-soil[.0. pag 616 library(agricolae) # Example 1 corr.0.test(block. Multivarial Analysis.analysis(cor.y)<-list(names.x<-correlation(x)$correlation path.x.x<.names) dimnames(corr.y.x.c(3.rbind(0.5.2)) corr.20)] y<-wilt[.x)<-list(names.92 PBIB.14] cor.analysis(corr.k.14.

s*r book<-design.2.4. design.7.7.4.6.4. de Mendiburu References 1.7. Wiley Classics Library Edition published 1992 See Also BIB. Value block Vector.R.7.6.6.7.4.5. 2.2.2.4.alpha Examples library(agricolae) library(MASS) # alpha design trt<-1:30 ntr<-length(trt) r<-2 k<-3 s<-10 obs<-ntr*r b <.4) dbook<-data.3) # dataset y<-c(5. lsd: least significant difference.1.PBIB.6.2.seed=5) book$book[.2.trt) .4.6.1.6.7.3.9.gl(20.5.2.6.4.1.test. Iterative Analysis of Generalizad Lattice Designs.1.6.r.test 93 Details Method of comparison treatment. tukey: Honestly significant differente.1.7. Experimental design. Statistics 19(1) 39-42.5.6.9. Cochran and Cox.1.6.k.1. consecutive numbers by replication trt Vector replication Vector y numeric vector k numeric constant method Character alpha Numeric Author(s) F.5.7.3]<. Williams (1977) Austral J.8.6.5.alpha(trt.2.3.5.frame(book$book. Second edition. 1.yield=y) rm(y.5.3. E.9.8.2.2.

replication.frequency=1.94 plot. other parameters of graphic Value x Numeric or list plot TRUE or FALSE counts Numeric frequency Numeric Author(s) Felipe de Mendiburu .graph. k=3) detach(dbook) # model$comparison # model$means plot.freq Histogram Description In many situations it has intervals of class defined with its respective frequencies.test(block.graph.freq(x. By means of this function. a object hist().counts=NULL.. nclass=NULL Arguments x a vector of values.PBIB. graph. 3=density plot logic nclass number of classes xlab x labels ylab y labels . Ojiva and to construct the table of complete frequency. yield. breaks="sturges". plot=TRUE. Usage plot.freq # analysis attach(dbook) model <.. polygon of frequency. the graphic of frequency is obtained and it is possible to superpose the normal distribution.graph. trt.freq() counts frequency and x is class intervals breaks a vector giving the breakpoints between histogram cells frequency 1=counts. 2=relative.

border="red". ylab="frequency".freq. intervals. xlab="YIELD" axis(1.graph.xlab="Variable". table.0. ylab="Frecuency") polygon..18.freq(h.col=NULL.freq.freq(h) print(measures) # frequency table full round(table.las=2) lines(density(yield).col="blue") grid(col="brown") #endgraph # Draw curve for Histogram h<.freq(h.freq.freq(h.counts=frequencies.freq.round(yield.col="red".03).col="yellow") axis(1.graph.0.7.axes=FALSE. only frecuency limits <-seq(10.5) frequencies <-c(2. frequency=3.2) #startgraph # Ojiva ojiva. xlab="Variable".seq(0. join.freq(limits. frequency=1.plot.1)) # summary.freq(limits.8.seq(0.freq.0. density=8.40.20.freq 95 See Also polygon.freq(h.freq(yield.axes=FALSE. ylab="Frequency") #endgraph # Statistics measures<-stat.col="bisque". stat.type="b". h1 <.h$breaks) axis(2.freq(h).freq(yield. frequency=2. ylab="f(yield)". col = "red".3. sturges.freq. col="blue".ylab="") title( main="Polygon of frequency".xlab="Classes") polygon.6.4) #startgraph h<-graph.freq.1)) # To reproduce histogram.ylab="Accumulated relative frequency") # only frequency polygon h<-graph. ojiva.0.4.col="red") title( main="Histogram and polygon of frequency".axes=FALSE.1) h<.h$breaks) axis(2.counts=frequencies.graph.subset(genxenv$YLD.genxenv$ENV==2) yield <.xlab=" ".col="yellow") axis(1.freq Examples library(agricolae) data(genxenv) yield <. normal.h$breaks) axis(2.graph.seq(0.border=FALSE. lwd = 2) title("Draw curve for Histogram") .

test(yield. A(b1).aov(yield ~ block + A + Error(plot)+ B + A:B.dfa. A(b2). Usage data(plots) Format A data frame with 18 observations on the following 5 variables. A(b3) comparison1 <-LSD.factor(plots[.(Ea +(b-1)*Eb)/(b*r) # Satterthwaite dfab<-(Ea +(b-1)*Eb)^2/(Ea^2/dfa +((b-1)*Eb)^2/dfb) # Comparison A.A.df.Eab) .dfab. block a numeric vector plot a factor with levels p1 p2 p3 p4 p5 p6 A a factor with levels a1 a2 B a factor with levels b1 b2 b3 yield a numeric vector Source International Potato Center.96 plots plots Data for an analysis in split-plot Description Experimental data in blocks. factor A in plots and factor B in sub-plots.residual(model$Within) Eb <-deviance(model$Within)/dfb Eab <. CIP Examples library(agricolae) data(plots) str(plots) plots[.Ea) comparison2 <-LSD.df. data=plots) summary(model) attach(plots) b<-nlevels(B) a<-nlevels(A) r<-nlevels(block) dfa <.1]) # split-plot analysis model <.residual(model$plot) Ea <-deviance(model$plot)/dfa dfb <.A[B=="b1"].test(yield[B=="b1"].1] <-as.

Genotype a factor with levels 102.Eab) comparison4 <-LSD.12 427. Usage data(plrv) Format A data frame with 504 observations on the following 6 variables.dfb.2 406.test(yield.B.28 157.11 235.7 450.test(yield[B=="b3"].6 319.B[A=="a2"].dfb. La Molina 02.B[A=="a1"].Eb) comparison7 <-LSD.dfab.Eab) # Comparison B.7 314.6 241. Huancayo.Eb) comparison6 <-LSD.plrv 97 comparison3 <-LSD.test(yield[A=="a1"].A[B=="b3"].A[B=="b2"]. B(a2) comparison5 <-LSD. San Ramon 02.dfab.18 104.26 163.21 402.2 255.15 346.20 320.7 405.3 506.22 121.dfb.2 Canchan Desiree Unica Locality a factor with levels Ayac Hyo-02 LM-02 LM-03 SR-02 SR-03 Rep a numeric vector WeightPlant a numeric vector WeightPlot a numeric vector Yield a numeric vector Source International Potato Center Lima-Peru References International Potato Center Lima-Peru Examples library(agricolae) data(plrv) str(plrv) . San Ramon 03. La Molina 03. B(a1).test(yield[B=="b2"].Eb) detach(plots) plrv Data clones from the PLRV population Description Six environments: Ayacucho.12 317.test(yield[A=="a2"].31 141.16 342.26 364.9 221.19 233.2 351.

. Usage polygon. normal.freq(height.col="red") #endgraph #startgraph h2<-graph. 3=density Author(s) Felipe de Mendiburu Delgado See Also polygon.xlim=c(6. Other parameters of the function hist Value histogram Object frecuency 1=counts.freq. join.98 polygon.freq. intervals. table.freq. . frequency=1.freq. It is necessary to execute the function previously hist.freq.col="red") #endgraph .border=FALSE.freq. counts(1).freq(histogram.freq(h1.frequency=2. graph..freq(h2. relative(2) and density(3) .frequency=1.frequency=2..col="yellow".freq Examples library(agricolae) data(growth) attach(growth) #startgraph h1<-hist(height.xlim=c(6. sturges.freq.freq Description The polygon is constructed single or on a histogram.) Arguments histogram Object constructed by the function hist frequency numeric. stat.freq The polygon of frequency on the histogram polygon..14)) polygon.14)) polygon. 2=relative.

Usage data(ralstonia) .solanacearum Description The assessment of the population of R. during days 15. 58. and 133 days after the infestation soil. References International Potato Center Examples library(agricolae) data(potato) str(potato) ralstonia Data of assessment of the population in the soil R. 29.solanacearum on the floor took place after 48 hours of infestation. More information on soil data(soil). date a numeric vector variety a factor with levels Canchan Unica harvest a numeric vector cutting a numeric vector Source Experimental data.potato potato 99 Data of cutting Description A study on the yield of two potato varieties performed at the CIP experimental station. 43. Usage data(potato) Format A data frame with 18 observations on the following 4 variables.

Sylvie Priou. q) . 2004. place a factor with levels Chmar Chz Cnt1 Cnt2 Cnt3 Hco1 Hco2 Hco3 Hyo1 Hyo2 Namora SR1 SR2 Day2 a numeric vector Day15 a numeric vector Day29 a numeric vector Day43 a numeric vector Day58 a numeric vector Day73 a numeric vector Day133 a numeric vector Details Logarithm average counts of colonies on plates containing half of M-SMSA 3 repetitions (3 plates by repetition) incubated at 30 degrees centigrade for 48 hours.Lima Peru. Data Kindly provided by Dr.100 random.ab(p. log(1+UFC/g soil) Source Experimental field. References International Potato Center.ab Randomizing a factorial in a block Description It generates a randomization in a block. Examples library(agricolae) data(ralstonia) str(ralstonia) random. CIP . Liliam Gutarra and Pedro Aley.ab Format A data frame with 13 observations on the following 8 variables. Usage random.

lsd.reg. x) Arguments trt treatment y dependent variable x independent variable Value trt factor y numeric x numeric . The objective is to find out if the linear models of each treatment come from the same population. y. design. There is a linear function in every treatment.3) Homologation of regressions reg.homog Description It makes the regressions homogeneity test for a group of treatments where each observation presents a linearly dependent reply from another one.nk Examples # factorial 2A3B in block library(agricolae) block.homog(trt. Usage reg.1 <-random. design.rcbd.ab(2.ab.homog 101 Arguments p level factor 1 q level factor 2 Value p numeric q numeric Author(s) Felipe de Mendiburu See Also design.crd. design. fact.

102 resampling.index) Resampling to find the optimal number of markers resampling. Chapman and Hall/CRC .production. npoints) Arguments A data frame or matrix of binary data size number of re-samplings npoints Number of points to consider the model Value A Matrix of numerical values of (1.cv(A. Tibshirani.cv Author(s) Felipe de Mendiburu References Book in Spanish: Metodos estadisticos para la investigacion.. Usage resampling. B. (1993) An Introduction to the Boostrap. R. size.homog(technology. Calzada Benza 1960 Examples library(agricolae) data(frijol) attach(frijol) evaluation<-reg. A method of the curvature.cv Description This process finds the curve of CV for a different number of markers which allows us to determine the number of optimal markers for a given relative variability.0) size constant numeric npoints numeric Author(s) Felipe de Mendiburu References Efron.

markers).expression(substitute(R^2==r.col="red".b) a+b/x x<-seq(2.cex=0.cv[.2).main=0.3))))) # Plot CV = a + b/n.a.1]) #endgraph resampling.list(r=round(Rsq.cv.cv) # Plot CV #startgraph limy<-max(table.cv[.length=50) y <.k) .b=round(coef[2]. ylab="CV".2)))) ) text(tx.expression(substitute(CV == a + frac(b.data.col="blue") #grid(col="brown") rug(table.cv(markers.model(model.8.markers fy<-function(x.cv[.8.model Resampling for linear models Description This process consists of finding the values of P-value by means of a re-sampling (permutation) process along with the values obtained by variance analysis.2)].data.tz.frame(study$table.main="Estimation of the number of markers") ty<-quantile(table.coef[1] + coef[2]/x lines(x.c(1.cv[.ty. list(a=round(coef[1].cv[.squared" table.max(table.cv[.cv <. similarity Examples library(agricolae) #example table of molecular markers data(markers) study<-resampling. cex=0.1]).0.cv[.model 103 See Also cv.estimate=py) print(table.npoints=25) # # Results of the model summary(study$model) coef<-coef(study$model) py<-predict(study$model) Rsq<-summary(study$model)$"r. ylim=c(10.resampling.as.2].2])+10 plot(table.frame=FALSE.xlab="number of markers".as. Usage resampling.1]) tz<-quantile(table.similarity.cex.95) text(tx.size=2.1) tx<-median(table.8.y.2].limy).

104 resampling. Tibshirani. R. (2001) Resampling Methods.model(model.k=100) .k=100) #example 4 Factorial in RCD data(potato) potato[.1]<-as.2]) model<-"cutting~variety + date + variety:date" analysis<-resampling.k=200) #example 2 Analysis of variance: RCD data(sweetpotato) model<-"yield~virus" analysis<-resampling.model Examples #example 1 Simple linear regression library(agricolae) data(clay) model<-"ralstonia ~ days" analysis<-resampling. B.factor(potato[. Boston .clay. (1993) An Introduction to the Boostrap.potato. Chapman and Hall/CRC Phillip I.Glycoalkaloids.factor(potato[.sweetpotato.2]<-as. Good.k=300) #example 3 Simple linear regression data(Glycoalkaloids) model<-"HPLC ~ spectrophotometer" analysis<-resampling.model(model.model Arguments model data k model in R data for the study of the model number of re-samplings Value model data k model data frame constant numeric Author(s) Felipe de Mendiburu References Efron.model(model..model(model. Berlin See Also simulation.1]) potato[. Basel . Birkhauser.

1 Measuring Soil Heterogeneity Source Statistical Procedures for Agriculture Research. 1976. 481. . Kwanchai A.rice 105 rice Data of Grain yield of rice variety IR8 Description The data correspond to the yield of rice variety IR8 (g/m2) for land uniformity studies. Gomez. V1 a numeric vector V2 a numeric vector V3 a numeric vector V4 a numeric vector V5 a numeric vector V6 a numeric vector V7 a numeric vector V8 a numeric vector V9 a numeric vector V10 a numeric vector V11 a numeric vector V12 a numeric vector V13 a numeric vector V14 a numeric vector V15 a numeric vector V16 a numeric vector V17 a numeric vector V18 a numeric vector Details Table 12. USA Pag. Usage data(rice) Format A data frame with 36 observations on the following 18 variables. Gomez and Arturo A. The growing area is 18x36 meters. Second Edition.

Gomez and Arturo A.farmer (9 levels) .) 2. Second Edition. babies: data.106 RioChillon References Statistical Procedures for Agriculture Research. mother: data.Lima Peru.) Details 1.clon (10 levels) . 2.frame: 30 obs. Evaluation of advanced Clones of potato in the Valley of Rio Chillon . USA Examples library(agricolae) data(rice) str(rice) RioChillon Data and analysis Mother and baby trials Description Mother/Baby Trials allow farmers and researchers to test best-bet technologies or new cultivars.block (3 levels) . Gomez.yield (kg. The "baby trials" with a subset of the treatments in the mother trial. CIP .clon (10 levels) . 3. .PERU (2004) Usage data(RioChillon) Format The format is list of 2: 1. of 3 variables: . References International Potato Center. Replicated researcher-managed "mother trials" with typically 10 treatments evaluated in small plots.frame: 90 obs. of 3 variables: . 1976. Source Experimental field. Unreplicated "baby trials" with 10 treatments evaluated in large plots. Kwanchai A.yield (kg.

yield.05. group=TRUE. The test is conservative. Multiple comparisons. DFerror. 18.test(yield. 4.model(model) attach(RioChillon$mother) comparison<-LSD. Fc. method is very general in that all possible contrasts can be tested for significance and confidence intervals can be constructed for the corresponding linear. main = NULL) Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degrees of freedom MSerror Mean Square Error Fc F Value alpha Significant level group TRUE or FALSE main Title Details It is necessary first makes a analysis of variance.test Description Scheffe 1959. Usage scheffe.922. trt.clon.test 107 Examples # Analisys the Mother/Baby Trial Design library(agricolae) data(RioChillon) # First analysis the Mother Trial Design. group=TRUE) # Third # The researcher makes use of data from both mother and baby trials and thereby obtains # information on suitability of new technologies or cultivars # for different agro-ecologies and acceptability to farmers.test(y.clon. group=TRUE) # Second analysis the babies Trial. model<-aov(yield ~ block + clon. attach(RioChillon$babies) comparison<-friedman(farmer. alpha = 0. scheffe scheffe.scheffe. RioChillon$mother) anova(model) cv. MSerror. .

group=TRUE.test Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric Fc Numeric alpha Numeric group Logic main Text Author(s) Felipe de Mendiburu References Robert O. SNK."virus". Duxbury. Dickey.test() attach(sweetpotato) df<-df.residual(model) MSerror<-deviance(model)/df Fc<-anova(model)["virus".err.scheffe. virus.test(yield. D. group=TRUE.J. main="Yield of sweetpotato\nDealt with different virus") # Old version scheffe. df. Steel.test.. Design of experiments. Fc. pp189 See Also HSD. main="Yield of sweetpotato. Dealt with different virus") detach(sweetpotato) . LSD. Kuehl. 2nd ed.group. bar.test.108 scheffe. data=sweetpotato) comparison <. bar. Torri.test .4] comparison <. copyright 2000. MSerror.test(model.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus.scheffe.duncan.(1997) Principles and Procedures of Statistics A Biometrical Approach. R.

col="blue") #endgraph .similarity.similarity 109 Matrix of similarity in binary data similarity Description It finds the similarity matrix of binary tables (1 and 0).cv Examples #example table of molecular markers library(agricolae) data(markers) distance<-similarity(markers) #startgraph tree<-hclust(distance.method="mcquitty") plot(tree. data binary Value A Numeric (0.1) Author(s) Felipe de Mendiburu See Also cv. resampling. Usage similarity(A) Arguments A Matrix.

k=100) #example 2 data(sweetpotato) model<-"yield~virus" .model Simulation of the linear model under normality Description This process consists of validating the variance analysis results using a simulation process of the experiment.model(model. The validation consists of comparing the calculated values of each source of variation of the simulated data with respect to the calculated values of the original data. since the P-Value is the probability of (F > F. then it is considered favorable and the probability reported by the ANOVA is accepted. categorical position of the columns of the data that correspond to categorical variables. k Number of simulations.model(model. file Data for the study of the model.file.model simulation.value). Usage simulation. categorical = NULL.model Examples library(agricolae) #example 1 data(clay) model<-"ralstonia ~ days" simulation.110 simulation. Value model Model file data frame categorical Numeric k constant numeric. Author(s) Felipe de Mendiburu See Also resampling. If in more than 50 percent of the cases they are higher than the real one.clay.k) Arguments model Model in R.

sinRepAmmi 111 simulation.potato.model(model. ENV a factor with levels A1 A2 A3 A4 A5 GEN a numeric vector YLD a numeric vector Source Experimental data.k=100) #example 4 data(potato) model<-"cutting~date+variety" simulation. References International Potato Center .Lima Peru.categorical=c(1. Examples library(agricolae) data(sinRepAmmi) str(sinRepAmmi) .model(model.sweetpotato.categorical=1. Usage data(sinRepAmmi) Format A data frame with 250 observations on the following 3 variables.k=100) #example 3 data(Glycoalkaloids) model<-"HPLC ~ spectrophotometer" simulation.Glycoalkaloids.3).k=100) sinRepAmmi Data for AMMI without repetition Description Data frame for AMMI analysis with 50 genotypes in 5 environments.model(model.2.

4. MSerror. The level by alpha default is 0.3595431.7) skewness(x) # value is 0.4. Tukey controls the error for all comparisons. but it is less conservative (finds more differences).05. DFerror.6.5. It is similar to SAS.05.4.test skewness Finding the skewness coefficient Description It returns the skewness of a distribution. where SNK only controls for comparisons under consideration. main = NULL) . alpha = 0.3. Usage skewness(x) Arguments x a numeric vector Value x The skewness of x See Also kurtosis Examples library(agricolae) x<-c(3.5.test(y.112 SNK. trt.NA. Usage SNK.test Student-Newman-Keuls (SNK) Description SNK is derived from Tukey. group=TRUE. is slightly asimetrica (positive) to the right SNK.2.

main="Yield of sweetpotato.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus."virus".SNK.SNK.test(model. HSD. Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric alpha Numeric group Logic main Text Author(s) Felipe de Mendiburu References Principles and procedures of statistics a biometrical approach Steel & Torry & Dickey.data=sweetpotato) comparison <. duncan.test.test . group=FALSE) # version old SNK. Dealt with different virus") SNK.test(model.test() attach(sweetpotato) . waller. Third Edition 1997 See Also LSD.test ."virus".test 113 Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degree free MSerror Mean Square Error alpha Significant level group TRUE or FALSE main Title Details It is necessary first makes a analysis of variance.

To which the different soil samples were dried to the environment.MSerror. 5 mm) and sterilized by steam 4 to 5 hours with a Lindinger Steam aerator SA150 and SA700. with the possible aim of eliminating bacteria saprophytic or antagonists to prevent the growth of bacteria (R. electrical conductivity CaCO3 a numeric vector MO a numeric vector CIC a numeric vector P a numeric vector K a numeric vector sand a numeric vector slime a numeric vector clay a numeric vector Ca a numeric vector Mg a numeric vector K2 a numeric vector Na a numeric vector Al_H a numeric vector . place a factor with levels Chmar Chz Cnt1 Cnt2 Cnt3 Hco1 Hco2 Hco3 Hyo1 Hyo2 Namora SR1 SR2 pH a numeric vector EC a numeric vector. plants.solanacearum). These analyses were conducted in the laboratory analysis of soils.SNK. screened (mesh 0. water and fertilizers in the La Molina National Agrarian University (UNALM).residual(model) MSerror<-deviance(model)/df comparison <.test(yield.114 soil df<-df. as full characterization of soil and special analysis of micro-elements.df. Usage data(soil) Format A data frame with 13 observations on the following 23 variables. group=TRUE) detach(sweetpotato) soil Data of soil analysis for 13 localities Description We analyzed the physical and chemical properties of different soils.virus.5xo.

Chz=Obraje-Carhuaz(Ancash).nonpar(data. Cnt2=Valle Dulce(Canete). Namora=Namora(Cajamarca). References International Potato Center . Hco1= Mayobamba-Chinchao(Huanuco). 2004.Lima.stability. ranking = FALSE) . Sylvie Priou. Liliam Gutarra and Pedro Aley. Hco3=San Marcos-Umari(Huanuco). variable = NULL.nonpar 115 K_Mg a numeric vector Ca_Mg a numeric vector B a numeric vector Cu a numeric vector Fe a numeric vector Mn a numeric vector Zn a numeric vector Details Cnt1= Canete. Cnt3=Valle Grande(Canete).nonpar Nonparametric stability analysis Description A method based on the statistical ranges of the study variable per environment for the stability analysis. Hyo1=La Victoria-Huancayo(Junin). Chmar=Chucmar-Chota(Huanuco. Hco2=Nueva IndependenciaChinchao(Huanuco). Data Kindly provided by Dr. Usage stability. Hyo1=El Tambo-Huancayo(Junin). Examples library(agricolae) data(soil) str(soil) stability. Sr2=La Chinchana-San Ramon(Junin). Source Experimental field. SR1= El Milagro-San Ramon(Junin). PERU.

frame(genotype=row.nonpar(haynes.]) cic.mean.5].nonpar Arguments data First column the genotypes following environment variable Name of variable ranking print ranking Value data Numeric variable Character ranking Logical Author(s) Felipe de Mendiburu References Haynes K G. pag 211-217.rm=TRUE)) means <-as.c(1.na.18)] data2<-CIC$oxapampa[.116 stability. Weingartner D P.nonpar(cic. Lambert D H.20)] cic <."YIELD".c(2. function(x) mean(x.7.7.c(1.6."greater") stability.6.mean<-data.data.by(cic[.1)]. 1998. Douches D S.na(cic.frame(means[.means) cic.mean) . See Also stability. Backlund J E. Phenotypic stability of resistance to late blight in potato clones evaluated at eight sites in the United States American Journal Potato Research 75.data2) attach(cic) means <.19.mean<-delete.ranking=TRUE) # Example 2 data(CIC) data1<-CIC$comas[. cic[.rbind(data1.par Examples library(agricolae) data(haynes) stability. Fry W and Stevenson W.names(means). Christ B J.17.

(iv) adjusts yield rank according to LSD (the adjusted rank labeled Y). (vii) sums adjusted yield rank. S.cov=NULL. Usage stability. the less stable the genotype). Arguments data rep MSerror alpha main cova name. Y.01.stability.05 and 0.file.cov Numeric Constant numeric Constant numeric Constant numeric Text FALSE or TRUE Text Vector numeric . and stability rating. and (viii) calculates mean YS(i) and identifies genotypes (selection) with YS(i) > mean YS(i).cova = FALSE.cov file.par stability. (v) determines significance of var(i) usign an aproximate F-test.rep.cov file. and 0 for nonsignificant var(i) ( the higher the var(i). (vi) assigns stability rating (S) as follows: -8.par 117 Stability analysis. (iii) calculates protected LSD for mean yield comparisons.name. The averages of the genotype through the different environment repetitions are required for the calculations.10 probability levels.MSerror. by rows the genotypes and columns the environment Number of repetitions Mean Square Error Label significant Title Covariable Name covariable Data covariable Details Stable (i) determines the contribution of each genotype to GE interaction by calculating var(i). for each genotype to determine YS(i) statistic. Value data rep MSerror alpha main cova name.par(data. SHUKLA’S STABILITY VARIANCE AND KANG’S Description This procedure calculates the stability variations as well as the statistics of selection for the yield and the stability. 0. -4 and -2 for var(i) significant at the 0. The mean square error must be calculated from the joint variance analysis.cov matrix of averages. (ii) assigns ranks to genotypes from highest to lowest yield receiving the rank of 1.main=NULL.1.alpha=0.

.5. 9.8.6.7.0) V11 <-c(5.2. 6. alpha=0. 6.5. 5. 4. file.3. 8.8.3. J.9.1. 7.6.c(9. 5.3.9.1900.6. 6.4.6. 8.7. 7.8.7.3.8.8. 8.118 stat.6.c(7.1. 9.4.6.6.8.0.6.V7. 8.4.5. 6. 7.7.1.1100.1. 9.4) V2 <.1. 9.2. Agron.7. 4. 7.7. 6.1.9.7.2.6. Simultaneous selection for yield and stability: Consequences for growers. 6.4.5. 7.4.0.c(10.7.3.frame(V1. 4. 3.9. 9. # replication rep= 4 # Mean square error.6.0.cov="Precipitation".4.nonpar Examples library(agricolae) # example 1 # Experimental data.6.9.8.5.4.3. 7.2.7.9.8.3. Manjit S.cov=precipitation) stat.4.7. 8. 5.4. MSerror=1. 8.1.3.1) V10 <-c(8.4.3. 6. cova=TRUE.8.0. 6. 5.8) V4 <.par(data. 6. name.2.5) data<-data.8.3. 6.4) V7 <.7.8.8.3.6. 3. 7.5.3.1.5.4.5.6.6.0. 6.9.2.9. Kang and Robert Mangari.8.8.2.6. 9. 4.3.6.3. 9.8.5.7.2. 4. MSerror = 1.0.5.8.8.8.5.8. 7.4.2100) stability.9.8.5. 5.2.5.1.6.7.6. main="Genotype") #example 2 covariable.6. 9.9. 6.1.c(7.9.7. J.6.9.V8. rep=4. 5. 7.4.freq Descriptive measures of grouped data .9.7.6) V12 <-c(8.1) V3 <.6.9.5.8.10.3.3.6.9. 8. 7.9. 8.7..3. 11. 5. 6. 6.6.4.V9.0.9.5.8.5. 9.4.9. 3.1.6.5. 85:754-757.5. 3.7.7.. Stable: A basic program for calculating stability and yield-stability statistics. 3.9.8.c(6. Agron.1. 4. 7.3) V9 <.2. 6.6.7. 17 # yield averages of 13 genotypes in localities V1 <.V12) rownames(data)<-LETTERS[1:17] stability.6.1800.4. 2.5.10. 2. 5.5. 7. 6.2.4.8.8.2. 5.9. 7.3.8.4) V5 <.6.V5. 7.7. 6. 6.5.8.7. main="Genotype".4.4.5.V10.1500.1. MSerror=1. 87:276-277 See Also stability.freq Author(s) Felipe de Mendiburu References Kang.9.7.8.8.9.2.4. 8. 5.2.2.6. 5. 8. 7. 8.8. 9. 7.3. 7.0. 7.1400.8. 8.5. 8.9.6.2. 6. 6.7. 4. 4.6.V3.8.9.7. 5. 7.7. 9.0.V11. 7.1. 1995.V4.8. 5.1300. 9. 5.3. 7. 7.2.1700. 8.9.3.8.2. M. 2.8.2.8.8. 1993.7.9. 7.9.V6.10.c(8. 10. 7.2. 4.7. alpha=0. 9.0. 7. 3.1.0. 5. 7. 4.5.8.1.6.8 # 12 environment # 17 genotype = 1.8. rep=4.9. S.11. 8. 7.0. 6.2.8.7) V6 <. 7. 8. 8.6.6.3.c(5.9.7.5.6) V8 <. 9.c(1000.8.1.1600.0.c(4.3.9.c(6.8.par(data.6.10. precipitation precipitation<.7.4.5.8.8.2000.8.0.V2.9. 7.9.1200.

freq(histogram) Arguments histogram Object create by function hist() Value histogram Object Author(s) Felipe de mendiburu See Also polygon. Usage stat.strip. ojiva. COL.freq. ROW. normal.freq(grouped) print(measures) strip. medium and mode of grouped data. intervals. table.freq. sturges.plot=FALSE) measures<-stat.freq. Usage strip.plot 119 Description By this process the variance and central measures ar found: average.freq. and the interaction analysis.plot(BLOCKS. the vertical-factor analysis.freq. Y) .freq. join.freq. graph.plot Strip-Plot analysis Description The variance analysis of a strip-plot design is divided into three parts: the horizontal-factor analysis.freq Examples library(agricolae) data(growth) attach(growth) grouped<-hist(height.

120

strip.plot

Arguments
BLOCKS

replications

COL

Factor column

ROW

Factor row

Y

Variable, response

Details
The strip-plot design is specifically suited for a two-factor experiment in which the desired precision
for measuring the interaction effects between the two factors is higher than that for measuring the
main efect two factors
Value
BLOCKS

vector, numeric

COL

vector, numeric

ROW

vector, numeric

Y

vector, numeric

Author(s)
Felipe de Mendiburu
References
Statistical procedures for agricultural research. Kwanchai A. Gomez, Arturo A. Gomez. Second
Edition. 1984.
Examples
# Yield
library(agricolae)
data(huasahuasi)
attach(huasahuasi)
YIELD<-huasahuasi$YIELD
attach(YIELD)
market <- y1da + y2da
non_market <- y3da
yield <- market + non_market
model<-strip.plot(block, clon, trt, yield)
comparison<-LSD.test(yield,clon,model$gl.a,model$Ea)
comparison<-LSD.test(yield,trt,model$gl.b,model$Eb)
# simple effects
out<-by(yield,trt, function(x) LSD.test(x, clon, model$gl.c, model$Ec))
out

sturges.freq

121

Class intervals for a histogram, the rule of Sturges

sturges.freq

Description
Number classes: k = 1 + 3.33 log10 (N).
Usage
sturges.freq(x)
Arguments
x

vector

Value
x

Numeric

Author(s)
Felipe de mendiburu
References
Reza A. Hoshmand. 1988. Statistical Methods for Agricultural Sciences, Timber Press, Incorporated, pag 18-21.
See Also
polygon.freq, table.freq, stat.freq, intervals.freq, graph.freq, join.freq,
ojiva.freq, normal.freq
Examples
library(agricolae)
data(natives)
attach(natives)
classes<-sturges.freq(size)
# information of the classes
breaks <- classes$breaks
breaks
#startgraph
# Histogram with the established classes
h<-graph.freq(size,breaks,frequency=1, col="yellow",axes=FALSE,
xlim=c(0,0.12),main="",xlab="",ylab="")
axis(1,breaks,las=2)
axis(2,seq(0,400,50),las=2)
title(main="Histogram of frequency\nSize of the tubercule of the Oca",

122

summary.graph.freq
xlab="Size of the oca", ylab="Frequency")
#endgraph

summary.graph.freq Frequency Table of a Histogram

Description
It finds the absolute, relative and accumulated frequencies with the class intervals defined from a
previously calculated histogram by "graph.freq".
Usage
summary.graph.freq(object,...)
Arguments
object

Object by function graph.freq()

...

other parameters of graphic

Value
object

Object by graph.freq()

Author(s)
Felipe de Mendiburu
See Also
polygon.freq, stat.freq, graph.freq, intervals.freq, sturges.freq, join.freq,
ojiva.freq, normal.freq
Examples
library(agricolae)
data(growth)
attach(growth)
h2<-graph.freq(height,plot=FALSE)
summary(h2)

The effect of two viruses (Spfmv and Spcsv) was studied. Usage data(sweetpotato) Format A data frame with 12 observations on the following 2 variables. FF (Spfmv) = Feathery mottle. The virus transmission was made in the cuttings and these were sown in the field. CIP . southern Peru. 50 sweetpotato plants were sown and 12 plots were employed. The treatments were the following: CC (Spcsv) = Sweetpotato chlorotic dwarf. FC (Spfmv y Spcsv) = Viral complex and OO (witness) healthy plants. Each treatment was made with 3 repetitions and at the end of the experiment the total weight in kilograms was evaluated. References International Potato Center.Lima Peru Examples library(agricolae) data(sweetpotato) str(sweetpotato) . In each plot. virus a factor with levels cc fc ff oo yield a numeric vector Source Experimental field.sweetpotato sweetpotato 123 Data of sweetpotato yield Description The data correspond to an experiment with costanero sweetpotato made at the locality of the Tacna department.

graph. ojiva.freq.freq.freq.freq Frequency Table of a Histogram table.freq.freq Description It finds the absolute. normal. relative and accumulated frequencies with the class intervals defined from a previously calculated histogram by the "hist" of R function. sturges.freq.freq.freq(h2) . intervals.freq Examples library(agricolae) data(growth) attach(growth) h2<-graph. stat.freq.freq() Value object Object by graph.freq(height. Usage table.plot=FALSE) table. join.124 table.freq() Author(s) Felipe de Mendiburu See Also polygon.freq(object) Arguments object Object by function graph.

stat(yield. The table must be organized by columns between variables and factors.virus..function(x) max(x)-min(x)) tapply.0.stat(yield.frame factors stat Method Value y Numeric x Numeric stat method = "mean".4)].mean) tapply.sweetpotato[.25.tapply.cotton[.75. grouped or crossed by factors.2].stat(yield.sd) tapply.c(1. .3.stat(y.function(x) quantile(x. Usage tapply.1].c(1.6)) detach(sweetpotato) # other case data(cotton) attach(cotton) tapply.frame variables x data.max) detach(cotton) # Height of pijuayo data(growth) attach(growth) . stat = "mean") Arguments y data.4)].cotton[.stat(yield. Author(s) Felipe de Mendiburu Examples library(agricolae) # case of 1 single factor data(sweetpotato) tapply.stat Description This process lies in finding statistics which consist of more than one variable.stat(sweetpotato[.mean) attach(sweetpotato) tapply.0.virus. x.stat(yield.6)-quantile(x.stat 125 Statistics of data grouped by factors tapply.virus..

2) vark(x.1.2.1.2.2.1.3. growth[. cor.2.1. Second Edition.3.y) .1.1.2.3.1.na.1. cor.mv Examples library(agricolae) x <-c(1.4.3.1. See Also cor.2.126 vark tapply.1.2.2.stat(height. function(x) mean(x.2.1.4.1. ties.matrix. Usage vark(x.1.2) y <-c(1. Kendall vark Description The Kendall method in order to find the K variance.1.3.1. y) Arguments x Vector y vector Details variance of K for Kendall’s tau Value x Numeric y Numeric Author(s) Felipe de Mendiburu References Numerical Recipes in C.2.4.4.vector.3.3.2:1].rm=TRUE)) detach(growth) Variance K.

group=TRUE. W.test 127 Multiple comparisons.waerden. alpha=0.test Description A nonparametric test for several independent samples. Conover.test(y. Value y Numeric trt factor alpha Numeric group Logic main text Author(s) Felipe de Mendiburu References Practical Nonparametrics Statistics.J. The van der Waerden (Normal Scores) waerden. 1999 See Also kruskal . trt. main=NULL) Arguments y Variable response trt Treatments alpha Significant level group TRUE or FALSE main Title Details The data consist of k samples of posibly unequal sample size.05. Usage waerden.

group=TRUE) waller Computations of Bayesian t-values for multiple comparisons Description A Bayes rule for the symmetric multiple comparisons problem.alpha=0. q. and 500.virus. Value K Numeric integer > 1. which roughly correspond for the two-level case to ALPHA levels of 0... . f.05.method. the procedure uses the default value of 100. and 0. By default.test(yield. 100. 100.method.test(observation.01.1. Fc) Arguments K Is the loss ratio between type I and type II error q Numerator Degrees of freedom f Denominator Degrees of freedom Fc F ratio from an analysis of variance Details K-RATIO (K): value specifies the Type 1/Type 2 error seriousness ratio for the Waller-Duncan test. examples 50. Reasonable values for KRATIO are 50.01.test(observation.group=TRUE) comparison<-waerden. 0. 500 q Numeric f Numeric Fc Numeric .128 waller Examples library(agricolae) # example 1 data(corn) attach(corn) comparison<-waerden. Usage waller(K.group=FALSE) # example 2 data(sweetpotato) attach(sweetpotato) comparison<-waerden.

Fc) }} W.test Examples # Table Duncan-Waller K=100. A Bayes Rule for the Symmetric Multiple Comparison Problem.D<-round(W. Waller-Duncan Description The Waller-Duncan k-ratio t test is performed on all main effect means in the MEANS statement. Journal of Statistical Computation and Simulation.100) f<-c(seq(4.waller. pages 1484-1504. DFerror.60. group=TRUE. Third Edition 1997. trt.D. Journal of the American Statistical Association 64.D) waller. R.D <-rep(0. and Kemp. See Also waller.16. A. (1976) Computations of Bayesian t-Values for Multiple Comparisons.30. E.10. Waller.20. f[j]. q[i]. Principles and procedures of statistics a biometrical approach Steel & Torry & Dickey.12.2).40. Usage waller.24.D)<-list(q. D. F=1.j]<-waller(K.test Multiple comparisons.40. R.2) dimnames(W.14. B. Fc. main = NULL) .D[i.m) for (i in 1:n) { for (j in 1:m) { W. See the K-RATIO option for information on controlling details of the test.20. (1969). 75. K. MSerror. A.n*m) dim(W. pages 169-172.f) print(W.test 129 Author(s) Felipe de Mendiburu References Waller. K = 100.test(y.2 q<-c(8.2 pag 649 Steel & Torry library(agricolae) K<-100 Fc<-1.120) n<-length(q) m<-length(f) W.D)<-c(n. and Duncan.

A Bayes Rule for the Symmetric Multiple Comparison Problem.01. and Duncan. (1969).1.130 waller. Third Edition 1997 Principles and procedures of statistics a biometrical approach .test Arguments y model(aov or lm) or answer of the experimental unit trt Constant( only y=model) or vector treatment applied to each experimental unit DFerror Degrees of freedom MSerror Mean Square Error Fc F Value K K-RATIO group TRUE or FALSE main Title Details It is necessary first makes a analysis of variance. Reasonable values for KRATIO are 50. which roughly correspond for the two-level case to ALPHA levels of 0. Journal of Statistical Computation and Simulation. and Kemp. 75. (1976) Computations of Bayesian t-Values for Multiple Comparisons. By default. pages 1484-1504. pages 169-172. R. A. D. Steel & Torry & Dickey. Waller. the procedure uses the default value of 100. E. A. R. 0. Value y class (aov or lm) or vector numeric trt constant (only y=model) or vector alfanumeric DFerror Numeric MSerror Numeric Fc Numeric K Numeric group Logic main Text Author(s) Felipe de Mendiburu References Waller. K. and 500. Journal of the American Statistical Association 64. 100. K-RATIO (K): value specifies the Type 1/Type 2 error seriousness ratio for the Waller-Duncan test. and 0.05. B.

residual(model) MSerror<-deviance(model)/df Fc<-anova(model)["virus".45).horiz=FALSE.test.test.xlim=c(0. virus. MSerror.xlim=c(0.waller.ylim=c(0. place a factor with levels Chmar Chz Cnt1 Cnt2 Cnt3 Hco1 Hco2 Hco3 Hyo1 Hyo2 Namora SR1 SR2 Day7 a numeric vector .waller.test() attach(sweetpotato) df<-df."virus".col="blue") bar.std=FALSE. Dealt with different virus") detach(sweetpotato) wilt Data of Bacterial Wilt (AUDPC) and soil Description Percentage of bacterial wilt and area under the curve of disease progression (AUDPC) relative tomato plants transplanted in different soil types artificially infested with R. df.45).group(comparison. main="Yield of sweetpotato. LSD.density=8.duncan. group=TRUE.group(comparison.err(comparison.test Examples library(agricolae) data(sweetpotato) model<-aov(yield~virus.wilt 131 See Also HSD.ylim=c(0.test .solanacearum 133 days before. SNK.err(comparison.col="green") #endgraph # Old version HSD.err. Fc.density=8.test(yield. bar. data=sweetpotato) comparison <.horiz=FALSE.4] comparison <.test(model.group.45).2)) # std = TRUE (default) is standard deviation # std = FALSE is standard error bar.col="red") bar.45).horiz=TRUE.std=FALSE. group=TRUE. main="Yield of sweetpotato\nDealt with different virus") #startgraph par(mfrow=c(2.horiz=TRUE.density=4. Usage data(wilt) Format A data frame with 13 observations on the following 15 variables. bar.density=4) bar.

days.Lima Peru.11.15.-1].35. Data Kindly provided by Dr. Liliam Gutarra and Pedro Aley.39. Days11 = evaluated to 11 days. 2004.132 wilt Day11 a numeric vector Day15 a numeric vector Day19 a numeric vector Day23 a numeric vector Day27 a numeric vector Day31 a numeric vector Day35 a numeric vector Day39 a numeric vector Day43 a numeric vector Day47 a numeric vector Day51 a numeric vector AUDPC a numeric vector relative a numeric vector Details Percentajes bacterial wilt.19. Day7 = evaluated to 7 days.type="relative") . see data(soil) and data(ralstonia) Source Experimental field.51) AUDPC<-audpc(wilt[.23. Sylvie Priou.days) relative<-audpc(wilt[. References International Potato Center.47.27.31. CIP .-1]. Examples library(agricolae) data(wilt) days<-c(7.43.

numeric(rownames(zz)) y<-as.10. Usage wxyz(model.5.zz. .wxyz 133 Completing missing data of a matrix according to a model wxyz Description It generates an information matrix z=f(x.x.5. x. y).z) # The response surface x<-as.7.y.2. phi = 30.theta = -20. cex=0.4. z) Arguments model obtained by ’lm’ x vector ’x’ y vector ’y’ z vector ’z’ relation of ’x’ e ’y’ Value model obtained by ’lm’ x Numeric y Numeric z Numeric Author(s) Felipe de Mendiburu See Also grid3p Examples library(agricolae) x<-c(1.y.2. col = 'green' .3) z<-c(4.NA.6.3.1) y<-c(2.3.3.2.shade= 0. y.NA) modelo<-lm(z~x+y) zz<-wxyz(modelo. It uses a linear model to complete the cells without information.nticks = 6.numeric(colnames(zz)) #startgraph persp(x.7.

matter dry foliage. ylab = 'Y'. which is a type of carbohydrate that can not be digested by the and the human body that have joined several beneficial properties in health. percentaje sucrose a numeric vector. zlab = 'Response') #endgraph yacon Data Yacon Description The yacon (Smallanthus sonchifolius) is a plant native to the Andes. locality a factor with levels. Oxapampa in PERU site a numeric vector dose a factor with levels F0 F150 F80 entry a factor with levels AKW5075 AMM5136 AMM5150 AMM5163 ARB5125 CLLUNC118 P1385 SAL136 replication a numeric vector. matter of dried roots. 0 to 1 . Usage data(yacon) Format A data frame with 432 observations on the following 19 variables. such as improve the absorption of calcium. grams dry a numeric vector. weight of fresh roots. dry matter kroner. grams roots a numeric vector. grams IH a numeric vector. weight of fresh foliage. considered a traditional crop in Peru and natural source of FOS. grams FOS a numeric vector. fructo-oligosaccharides.xlab = 'X'. replications height a numeric vector. centimeters stalks a numeric vector.134 yacon ticktype = 'detailed'. Cajamarca. plant height. Index harvest. reducing the level of triglycerides and cholesterol and stimulate better gastrointestinal function. grams wfk a numeric vector. degrees Brix foliage a numeric vector. grams wff a numeric vector. Lima. percentaje glucose a numeric vector. percentaje fructose a numeric vector. number of stalks wfr a numeric vector. weight fresh kroner. percentaje brix a numeric vector.

weight = 100g and frijol = 50g then. Source CIP. References International Potato Center. Plant of frijol. Experimental field.5 or 50 percentaje.Lima Peru. root) on dry basis. Examples library(agricolae) data(yacon) str(yacon) . 2003. Part usable in a proportion of total mass dissected. Degrees Brix is a measurement of the mass ratio of dissolved sugar to water in a liquid. Data Kindly provided by Ivan Manrique and Carolina Tasso. fruit. CIP .yacon 135 Details Proportion or fraction of the plant that is used (seeds. IH = 50/100 = 0.

75 scheffe. 82 ojiva. 104 SNK. 95 ∗Topic cluster consensus. 108 soil. 51 Glycoalkaloids. 93 plrv. 19 cotton. 125 ∗Topic htest duncan.Index ∗Topic aplot AMMI.lsd. 28 design. 131 ∗Topic design design.freq. 77 melon. 119 table. 63 LSD. 6 136 . 14 CIC.crd. 59 heterosis.split.test. 78 natives. 16 ComasOxapampa.test.err.freq. 57 haynes. 86 paracsho.graph. 97 ∗Topic distribution summary. 60 hgroups. 121 waller.strip. 50 genxenv. 67 random. 91 polygon. 109 waller. 9 graph.bib.freq. 58 Hco2006. 43 fact.test. 41 design. 17 hcut. 48 index.test. 61 huasahuasi. 31 design.contour. 39 design. 33 design.alpha. 36 design. 94 potato.freq. 16 corn. 96 ralstonia. 52 normal. 22 DC. 38 design.ab.freq. 15 clay.nk. 102 RioChillon. 26 disease. 65 LxT. 44 frijol. 96 rice. 87 plots. 111 sweetpotato.group. 76 markers.graph. 45 HSD.graeco. 8 bar. 120 wilt.freq.test. 62 ∗Topic datasets Chz2006. 40 design. 126 ∗Topic manip audpc. 128 yacon. 55 growth. 103 sinRepAmmi. 35 design. 54 greenhouse. 83 plot. 80 pamCIP.ab.smith. 29 design.freq.rcbd.dau.lattice. 52 grass. 5 bar. 32 design.test.cyclic.

107 stability. 74 AMMI. 46. 48.split. 37. 48.test.test. 45. 49 kendall. 34. 33. 9. 39.lattice. 109 137 stat. 76.model. 70 kruskal. 22 cv. 26. 16 ComasOxapampa. 6 bar.cyclic. 8. 98 design. 30. 39. 29. 110. 38. 30. 2 ∗Topic regression reg.ab. 22 correlation. 116 ∗Topic multivariate correl. 32. 98 design. 41–43. 47. 98 design. 130 ∗Topic math grid3p.na. 2 BIB.test. 36 design.stat. 63 corn.freq. 11. 21. 76. 48.na.freq. 12 Chz2006. 90 design.contour. 46 friedman. 43 disease. 29. 73 montecarlo. 128 bar. 48. 28. 79 design. 2. 12 DAU. 32.strip. 124 ∗Topic optimize resampling. 106 DAU. 10. 37–39. 106 simulation. 41–43.model. 2 agricolae-package. 44 duncan. 27 lastC. 33. 76. 41 design. 98 ∗Topic univar cv. 98 design. 6.test.rcbd.lsd.freq. 123 waerden. 114 strip. 105.dau. 37–39. 36. 99 ∗Topic package agricolae-package. 34. 64.cv. 48. 12. 61.group. 128 BIB.test. 79 order.analysis. 105. 115 tapply. 32. 71. 11 carolina. 25 lineXtester. 31.test.err.bio. 72 fact. 26 delete. 23 index. 20 correlation. 32.crd. 56 ∗Topic models AMMI. 19 correl. 2 AMMI.homog.test. 15 clay. 20. 118 wxyz. 16 consensus. 5 audpc. 100. 27 density.bib. 89 cotton.model. 112 vark.similarity. 33 design.nonpar.test. 50. 90 carolina.INDEX delete. 38 design. 35 design. 122 agricolae (agricolae-package).par. 89 similarity. 21. 85 sturges. 34 design. 68 join. 29. 38.graeco. 29.similarity.test. 88 resampling. 98 . 73 nonadditivity. 30.nk. 40. 100 ∗Topic nonparametric durbin.group. 41–43. 81 PBIB. 69 kurtosis. 29.stat. 71 stability. 72 skewness. 66 intervals.model. 24 path.plot.alpha. 23 cv. 32. 13. 33.freq. 14 CIC. 21 cv. 128 durbin. 105. 17. 24. 9. 33. 84 order. 25 DC.

113. 46. 61 hgroups. 121 summary.nonpar. 100.freq. 118. 92. 64. 53. 60. 128 soil. 73 lineXtester. 104 similarity.smith. 26. 46. 69. 124 kurtosis.test. 95. 76.graph. 69.test. 46. 18. 118. 103 lastC. 119. 63. 116. 70. 118. 72. 92. 10. 84. 121 strip.stat. 70. 69. 111 stability. 77 melon. 24. 54 greenhouse. 9. 123 join. 85 genxenv.cv. 76.freq. 83. 107 rice. 82.graph. 114 stat. 86 paracsho. 65 index. 61. 10. 10. 88 PBIB. 128 huasahuasi. 116. 119. 98 resampling. 83. 105. 95. 108 skewness. 58 hclust.analysis. 101. 95. 70. 50 order. 96 random. 112. 119. 95.freq.test. 119. 83. 102 RioChillon. 75. 109 SNK. 53. 82. 109. 82. 24. 87 path. 99. 47. 83. 73 LSD. 79 table. 116. 9.model. 69.138 INDEX friedman. 82. 55 grid3p. 121 grass.ab.model. 70. 69. 72 frijol. 120 markers. 125 waller. 121 potato. 121 waerden. 63 Hco2006. 126. 106 simulation. 69. 24. 66 index. 92. 70 kruskal. 37–39. 126 wilt. 93 plrv. 119. 82.freq. 80 nonadditivity. 119. 68. 116.test.freq. 82.bio. 67 intervals. 128 .test. 24.freq. 83. 53. 72. 57 pamCIP. 82. 119. 119 sweetpotato. 9. 69. 70.freq. 46. 78 montecarlo. 115 stability. 119. 121 kendall. 73. 82. 9. 105. 76. 130 growth. 106 resampling. 116 sturges. 59 hcut. 92.test. 3. 95. 53. 81 normal. 92. 63 heterosis. 83. 52 graph.par. 47. 32. 52. 26. 70. 69. 110. 26. 118.freq. 122 natives. 50. 70. 24. 116. 83. 48.freq. 118. 94 polygon. 83. 118. 118.freq. 10. 97 reg. 64. 116. 116. 95. 118.test.homog. 70. 76 scheffe. 62 HSD. 29. 53. 18. 10. 110. 83. 53. 91 plots. 53. 61. 89 plot. 121 tapply.freq. 64. 110. 95. 115. 92. 41–43. 95. 71. 109 ojiva. 18. 56. 121 vark. 49.plot. 121 ralstonia. 124 waller. 107 sinRepAmmi. 116.stat. 95. 51 Glycoalkaloids. 82. 96 haynes. 53. 69.group. 85. 105. 92. 100. 128 LxT. 86 order. 92. 118.

130 yacon. 131 139 . 57.INDEX wxyz.