You are on page 1of 42

STATISTICS :

INTRODUCTION AND ITS


USE IN MANAGERIAL
DECISION MAKING
(CHAPTER – 1)

- GAURAV GOYAL
(LECTURER)

03/10/10 1
CONTENTS
 INTRODUCTION
 BROAD CLASSIFICATION
 WHAT IS STATISTICS?
 CHARACTERISTICS OF STATISTICAL
DATA
 STATISTICAL METHODS:CLASSIFICATION
 OPERATION RESEARCH MODELS
 SOME STATISTICAL TECHNIQUES
 ADVANTAGES OF QT TO
MANAGEMENT
 QT IN BUSINESS AND MANAGEMENT
 03/10/10 2
INTRODUCTION
BEFORE INDUSTRIAL REVOLUTION AFTER INDUSTRIAL REVOLUTION
ision based on past experience
Decisionand
making
intuition
got complicated and quantitative techn

REASONS
1. Complexity of today’s managerial activities
em since customers 2were
. Availability of owner
known to the different
. type of tools for quantitative analysis of
3. Availability
eded since choice and of customer
requirement of high speed
wascomputers
known. to apply the quantitative techni
uired since manager used to work with the workers at the shop floor level.
t required since progress was recorded daily at the work centre.
uired by the owner for decision making were known to him.

03/10/10 3
BROAD CLASSIFICATION
QUANTITATIVE TECHNIQUES

STATISTICAL TECHNIQUES
OPERATION RESEARCH (OR PROGRAMMING) TECHNIQU

03/10/10 4
WHAT IS STATISTICS?

MISCONCEPTION
tion, analysisPeople
, interpretation and presentation
sometimes refer of of
it as a group numerical
data butdata .
actually it

APPLICATION
nce of statistics
Provides
is based
tools
on for
mathematical
predictionthought
and forecasting
and derivation
using. data a

03/10/10 5
CHARACTERISTICS OF STATISTICAL
DATA
 Must be aggregate of facts
 Should be affected to a marked
extent by multiplicity of causes
 Must be enumerated / estimated to
a certain standard of accuracy.
 Must have been collected in a
systematic manner for a pre-
determined purpose.
 Must be placed in relation to each
other
 Must be numerically expressed.
03/10/10 6
STATISTICAL
METHODS:CLASSIFICATION
STATISTICAL METHODS

DESCRIPTIVE STATISTICS INDUCTIVE STATISTICS STATISTICAL DECISION THEORY

1.Data collection 1.Statistical Inference1. Analysis of Business Decision


2.Presentation 2.Estimation

03/10/10 7
CONTINUED……

collection of data.

s to summarize a class’s examination effort and uses those to reach conclusions ab

APPLIED STATISTICS

ount for randomness and uncertainty and then used to draw inferences about the po

03/10/10 8
DESCRIPTIVE
STATISTICS(EXAMPLE)

03/10/10 9
INFERENTIAL STATISTICS (EXAMPLE)

lbs manufactured with the new


A sample
filament
of . 200
Average
bulbs lifetime
is manufactured
is unknown
with
. the new filament .

rage is used to estimate


The sample
the population
data provide
average
a sample
. average lifetime of 76 hours per

03/10/10 10
STATISTICAL DECISION THOERY
 Deals with analyzing complex
business problems with alternative
courses of action and possible
consequences.
 Relies on:-
 a) nature of problem
 b) decision environment
STATE OF DECISION CONSEQUENCES

Certainty Deterministic
Risk Probabilistic
Uncertainty Unknown
Conflict Influenced by opponent

03/10/10 11
OPERATION RESEARCH
MODELS
 MODELS BASED ON PURPOSE

DESCRIPTIVE MODEL EXPLANATORY MODEL

be the behavior
1.Used
of atosystem
explain
based
behavior
on certain
of system
information
by establishing
. relationship bet
roblem situation more vividly including the alternative choices.

for
explain
a inventory
variations
iteminfor
productivity
a stated period
by establishing
by keeping relation
record ofamong
demand
thelevels
factors
andsut

03/10/10 12
OPERATION RESEARCH
MODELS
 MODELS BASED ON PURPOSE

PREDICTIVE MODEL PRESCRIPTIVE MODEL

redict status
1.Provides
of a system
norms for
in the
comparison
near future
of alternative
based on data
solutions
. to select the be

redict stock prices for any given level of earning per share (EPS).

03/10/10 13
OPERATION RESEARCH
MODELS
 MODELS BASED ON DEGREE OF ABSTRACTION

PHYSICAL MODEL GRAPHIC MODEL SCHEMATIC MODEL

ooks like the 1real


.A model
thing
showing
, eitherrelationship
reduced in size
in
1.A amodel
or
graphical
scaled
depicting
up
manner
. sequence
. of activ
2 . MORE ABSTRACT than the Physical2 . MOREmodel
ABSTRACT
. than the Graphical

ng maps, plant layout


e.g. acharts
flow , chart
airplane
e.g., organisation
computer
model program
etc. chartwhose main features are depic

03/10/10 14
OPERATION RESEARCH
MODELS
 MODELS BASED ON DEGREE OF ABSTRACTION

ANALOG MODEL MATHEMATICAL MODEL

ith physical models, and are


1.Represents
not replica
the of
systems
the problem
by using
situations
mathematical
. symbols a
, works like the system it2 .represents
MOST ABSTRACT
.
n schematic.

e.g. children toys, model rail-roads


e.g. the
etc. input-output model of the national eco

03/10/10 15
OPERATION RESEARCH
MODELS
 MODELS BASED ON DEGREE OF CERTAINTY

DETERMINISTIC MODEL PROBABILISTIC MODEL

ion of each course1.of


Situations
action results
in which
ineach
unique
course
and known
of action
pay-off
can. result in more

near programming, transportation and assignment


e.g. simulation
models. model, decision-theory model

03/10/10 16
OPERATION RESEARCH
 MODELS
MODELS BASED ON SPECIFIC BEHAVIOR
CHARACTERISTICS
 STATIC MODELS DYNAMIC MODEL

icular
on-makerset
hasoftofixed
make conditions
a sequence and
of optimal
do not change
decisions
in aatshort
every
-term
decision
period.point rega
only
on changes
one decision
over a required
period offor
timea. period.

urces for a particular product, Linear programming


e.g. product etc
development, dynamic programmi

03/10/10 17
Link for Excel

http://www.free-training-tutorial.com/format-cells.htm
 

03/10/10 18
03/10/10 19
OPERATION RESEARCH
 MODELS
MODELS BASED ON SPECIFIC BEHAVIOR
CHARACTERISTICS
 LINEAR MODELS NON - LINEAR MODELS

in which each component 1exhibits


.If one or
a linear
more components
behavior. in a model exhibits a non-line

e.g. e.g.
Z = 5 + 3x Z = 5x2 + 3xy + y2

03/10/10 20
OPERATION RESEARCH
 MODELS
MODELS BASED ON PROCEDURE OF SOLUTION


ANALYTICAL MODELS SIMULATION MODELS

and1.Is
solved
the experimentation
by known mathematical
on a mathematical
or analytical
structure
techniques
of real
to yield
life asystem
general
. sol
2.Done by inserting specific values of decision variables into the given stru

ar programming, game
e.g. theory
test effect
modelsof
, inventory
differentcontrol
number models
of service
etc counters based on

03/10/10 21
Measures of Location
■Mean

■Median If the measures are computed


for data from a sample,
■Mode they are called sample statistics.
■Percentiles

■Quartiles If the measures are computed


for data from a population,
they are called population parameters.

A sample statistic is referred to


as the point estimator of the
corresponding population parameter.

03/10/10 22
Mean
 The mean of a data set is the
average of all the data values.
x
 The sample mean is the point
estimator of the population mean
µ .

03/10/10 23
Sample Mean x

Sum
Sum ofof the
the values
values
of
of the
the nn observations
observations

∑x i
x=
n

Number
Number ofof
observations
observations
in
in the
the sample
sample

03/10/10 24
Population Mean µ

Sum
Sum ofof the
the values
values
of
of the
the N
N observations
observations

∑x i
µ=
N

Number
Number ofof
observations
observations in
in
the
the population
population

03/10/10 25
Sample Mean

n Example: Apartment Rents


Seventy efficiency apartments were randomly
sampled in a small college town. The monthly rent
prices for these apartments are listed on the next slide.

03/10/10 26
Sample Mean

n Apartment Rent Sample Data


445 615 430 590 435 600 460 600 440 615
440 440 440 525 425 445 575 445 450 450
465 450 525 450 450 460 435 460 465 480
450 470 490 472 475 475 500 480 570 465
600 485 580 470 490 500 549 500 500 480
570 515 450 445 525 535 475 550 480 510
510 575 490 435 600 435 445 435 430 440

03/10/10 27
Sample Mean

x=
∑ x i
=
34,356
=490.80
n 70
445 615 430 590 435 600 460 600 440 615
440 440 440 525 425 445 575 445 450 450
465 450 525 450 450 460 435 460 465 480
450 470 490 472 475 475 500 480 570 465
600 485 580 470 490 500 549 500 500 480
570 515 450 445 525 535 475 550 480 510
510 575 490 435 600 435 445 435 430 440

03/10/10 28
Median
n The median of a data set is the value in the middle
when the data items are arranged in ascending order.
n Whenever a data set has extreme values, the median
is the preferred measure of central location.
n The median is the measure of location most often
reported for annual income and property value data.
n A few extremely large incomes or property values
can inflate the mean.

03/10/10 29
Median

n For an odd number of observations:

26 18 27 12 14 27 19 7 observations

12 14 18 19 26 27 27 in ascending order

the median is the middle value.

Median = 19

03/10/10 30
Median

n For an even number of observations:

26 18 27 12 14 27 30 19 8 observations

12 14 18 19 26 27 27 30 in ascending order

the median is the average of the middle two values.

Median = (19 + 26)/2 = 22.5

03/10/10 31
Median
Averaging the 35th and 36th data values:
Median = (475 + 475)/2 = 475
425 430 430 435 435 435 435 435 440 440
440 440 440 445 445 445 445 445 450 450
450 450 450 450 450 460 460 460 465 465
465 470 470 472 475 475 475 480 480 480
480 485 490 490 490 500 500 500 500 510
510 515 525 525 525 535 549 550 570 570
575 575 580 590 600 600 600 600 615 615
Note: Data is in ascending order.

03/10/10 32
Mode
n The mode of a data set is the value that occurs with
greatest frequency.
n The greatest frequency can occur at two or more
different values.
n If the data have exactly two modes, the data are
bimodal.
n If the data have more than two modes, the data are
multimodal.

03/10/10 33
SOME STATISTICAL
TECHNIQUES
 MEASURES OF CENTRAL TENDENCY
1.MEAN: Computed by dividing sum of
the values of observations by the
number of items.
2.MEDIAN: Item which lies exactly
midway between the lowest and
the highest value when data is
arranged in ascending or
descending order.
3.MODE: It is the central value that
occurs most frequently.
03/10/10 34
SOME STATISTICAL
TECHNIQUES
 MEASURES OF DISPERSION

 The difference between the
maximum and the minimum value
of a distribution is called the
measure of dispersion.
1.SKEWNESS: the measure of the
direction and degree of symmetry
are called measures of skewness
2.KURTOSIS: The measure of
peakedness in a distribution is
called Kurtosis.
03/10/10 35
SKEWNESS AND KURTOSIS

03/10/10 36
SOME STATISTICAL
TECHNIQUES
 CORRELATION
àIt measures the degree to which the
change in the dependent variable
is associated with change in the
independent variable.
 REGRESSION ANALYSIS
àUsed for determining the causal
relationship between 2 variables.
 TIME SERIES ANALYSIS
àConsists of a set of data recorded
over successive periods of time.
àMeasures the effect of the following
03/10/10 37
SOME STATISTICAL
TECHNIQUES
 TIME SERIES ANALYSIS
àConsists of a set of data recorded
over successive periods of time.
àAnalysis is done on the above data
taking the periodic changes,
cyclical changes, seasonal
variations and random variations
into consideration.

03/10/10 38
ADVANTAGES OF QT TO
MANAGEMENT
 Definiteness
 Condensation
 Comparison
 Formulation of policies
 Formulating and testing hypothesis
 Prediction

03/10/10 39
QT IN BUSINESS AND
MANAGEMENT
 MANAGEMENT
 i) Marketing:
 analysis market research
information
 Statistical records for building
and maintaining an extensive market
 Sales forecasting
 ii) Production
 Production planning, control and
analysis
 Evaluation of machine
performance 03/10/10 40
QT IN BUSINESS AND
MANAGEMENT
 iii) Finance, Accounting and
Investment:
 financial forecast, budget
preparation
 financial investment decisions
 Selection of securities
 Auditing function
 iv) Personnel
 labour turnover rate
 employment trends
 performance appraisal
 wage rates and incentive plans
03/10/10 41
QT IN BUSINESS AND
MANAGEMENT
 ECONOMICS
 measure of GNP.
 determination of business cycle.
 comparison of market prices
etc.
 analysis of population
 formulation of appropriate
economic policies
 RESEARCH AND DEVELOPMENT
 development of new product
lines
 optimum use of resources
03/10/10 42