132 BUSINESS STATISTICS: PROBLEMS AND SOLUTIONS

the population (in per cenu

preceding toss. Increasein in the USA.
of successive tosses of a coin are independent t of its production (in per cent) per year
per year in India is independent of increase in
wheat

Compound Events their simultaneous OCCUTrence js

with each other, then dependent.
When two or more events Occur 1n connection
independent or (ii)
called a compound event. These events may be (i)
That is, one of them
Equally Likely Events
each has an equal chance to occur. may be
TIwo or more events are said to be equallylikely if each number expected to
in preference to the other. For example, long run.
cannot be expected to occur of times in the
uppermost face of a rolling die the same number
O ne
Gomplementary Events E-bar) contains all the
complement denoted by (read as
If Eis any subset of the sample space, then its sample space, then
elements of the sample space that are not part of E. If S denotes the
elements not in E}.
E =S- E ={AIl sample lakh, written as E =
represents companies with sales less than or equal to Rs. 25events must be mutually
For example, if E complement of the set E = fx:x> 25}. Obviously such
:xS 25 }, then this set is a
exclusive and collective exhaustive.

5.3 DEFINITION OF PROBABILITY

measure (between 0 and 1
general definition of probability states that probability is a numerical ol
A of an uncertain event. Conceptual approaches
inclusively) of the likelihood or chance of occurrence
calculating the probability of an event are as follows :
Classical Approach possible outcomes (finite in number) of an experiment
This approach is based on the assumption that all the that, during a random experiment, if there are 2
are mutually exclusive and equally likely. It states and "b"possible outcomes where the event Adoes
possible outcomes where the favourableevent A occurs the
not occur, and all these possible outcomes are mutually exclusive, exhaustive, and equiprobable, then
defined as
probability that event A will occur is
Number of favourable outcomes qA)
PA) = a+h Total number of possible outcomes (S)
equally likely
llustrations If a fair dice is rolled, then on any trial, each event (face or number) is
and theretoi
occur Since there are six equally likely exhaustive events, each will occur l/6 of the time,
oCcurring is 1/6.
the probability of any one event events,the
In general, for a random experiment with n mutually exclusive, exhaustive, equiprobable
probability of any of the events is equal to l/n.
Relative Frequency Approach numberof
This approach is based on the assumption that a random experiment can be repeated alarge a
random
times under identical conditions where trials are independent to each other. While conductingmany times
experiment we may or may not observve the desired event. But as the experiment is repeated
ofthetime(e
that event may OCcur some proportion of time. Thus, the approach calculates the proportion7
therolntie freguency) with which the event oCCurs over anintinitenumber ofrepetitions of theexperiment
identicalconditions.
t)
or
4,5,
Illustrations If a dice is tossed ntimes and s denotes the number of times event A(i.e., number areal
and
occurs, then the ratio PA) = dsln gives the proportion of times event A OCcurs in ntrials. change afe
called relative frequencies of the event in n trials. Although our estimate about PA) may valueasth
every trial, yet we will find that the proportion qs)/n tends to cluster around a unique cenral
 FUNDAMENTALS OF PROBABILITY 133
number of trials n becomes even larger. This
unique central value (also called
defined as probability of event A) is
PA) = lim
n-’o
.bere ds) represents the number of times that an
event s occurs in n trials of an
experiment.
Subjective Approach
This subjective approach of calculating
and experience concerning the probability is always based on the degree of beliefs,
likelihood of
individual's beliefs, assessment, and judgmentoccurrence of a random event. It is thus a way to convictions,
about a random phenomenon. Probability quantify an
ocurrence of an event may be based on just guess or on having some idea about the assigned for the
nast occurrences of the event. This relative
approach must be used when either sufficient data are notfrequency of
sources of information giving different results are not. known. available or
 Probability Distributions 6
6.1 INTRODUCTION

AIisung of all the possible outc omes of a random sat lable wth cah outcone s aMK laled pt obabaln
xuTTenceiscalled prtaabalts dsrabutum lhe numeal valuc ot a tandom vat Labie depends upon t
outcome of an expernent and mav be datteret tor dttetent trab ot the ane xperment Ihe s d al
such values so obtauned is called the tatige yueot the landon at Labie
lllustration: If acon is tossed twKe. then the sanple spauc ot ens for thas random experment
(HH. I H HI ID
In this case, if the deasofh- thaket s inteited uo khom the prvt
satadst dastrstuton for the number
heads on (wo tosses o the co) then a tandes11afLale may be detncd as

The values ofx wl dcpend os hance and a takr the valucs

IT= 0 Thus the tange space ot H
H2. HI=I. TH = l
n (0 1 2