STATISTICS

CA FOUNDATION - PAPER 3 - BUSINESS MATHEMATICS,

LOGICAL REASONING AND STATISTICS

This capsule on Foundation - Paper 3 - Business Mathematics, Logical Reasoning and Statistics , will enable the students
to understand and apply the techniques of developing discrete and continuous probability distributions.

Chapter 17 : Theoretical Distributions

In this chapter we will discuss the probability theory by Binomial Distribution

considering a concept and analogous to the idea of frequency One of the most important and frequently used discrete probability
distribution. In frequency distribution where to the total distribution is Binomial Distribution. It is derived from a particular
frequency to different class intervals, the total probability type of random experiment known as Bernoulli process named
(i.e. one) is distributed to different mass points is known as after the famous mathematician Bernoulli. Noting that a 'trial' is an
theoretical probability distributions. attempt to produce a particular outcome which is neither certain
• Discrete Random variable. nor impossible, the characteristics of Bernoulli trials are stated
• Continues Random variable. below:
(i) Each trial is associated with two mutually exclusive and
exhaustive outcomes, the occurrence of one of which is known
as a 'success' and as such its non occurrence as a 'failure'. As an
Importance of theoretical probability distribution.:
example, when a coin is tossed, usually occurrence of a head is
(a) An observed frequency distribution, in many a case, may
known as a success and its non–occurrence i.e. occurrence of a
be regarded as a sample i.e. a representative part of a large,
tail is known as a failure.
unknown, boundless universe or population and we may be
(ii) The trials are independent.
interested to know the form of such a distribution.
For Example: By fitting a theoretical probability distribution.
• Length of life of the lamps produced by manufacturer
up to a reasonable degree of accuracy. We may note the following important points in connection with
• The effect of a particular type of missiles, it may be binomial distribution:
possible for our scientist to suggest the number of such (a) As n >0, p, q ≥ 0, it follows that f(x) ≥ 0 for every x
missiles necessary to destroy an army position. Also ∑ f(x) = f(0) + f(1) + f(2) + …..+ f(n) = 1
• By knowing the distribution of smokers, a social activist (b) Binomial distribution is known as biparametric distribution as
may warn the people of a locality about the nuisance of it is characterised by two parameters n and p. This means that if
active and passive smoking and so on. the values of n and p are known, then the distribution is known
x
completely.
(b) Theoretical probability distribution may be profitably
employed to make short term projections for the future. (c) The mean of the binomial distribution is given by µ = np
(c) Statistical analysis is possible only on the basis of theoretical (d) Depending on the values of the two parameters, binomial
probability distribution. distribution may be unimodal or bi- modal., the mode of
A probability distribution also possesses all the characteristics binomial distribution, is given by µ0 =the largest integer
of an observed distribution. We define population mean, contained in (n+1)p
population median, population mode, population standard if (n+1)p is a non-integer (n+1)p and (n+1)p - 1 if (n+1)p is an
deviation etc. exactly same way we have done earlier. These integer
characteristics are known as population parameters. (e) The variance of the binomial distribution is given by
σ2 = npq
Since p and q are numerically less than or equal to 1, npq < np
A probability distribution may be either a discrete probability variance of a binomial variable is always less than its mean.
distribution or a Continuous probability distribution depending Also variance of X attains its maximum value at p = q = 0.5 and
on the random variable under study. this maximum value is n/4.
(f ) Additive property of binomial distribution.
If X and Y are two independent variables such that
X~ β (n1, P)
Two important discrete probability distributions
and Y ~ β (n2, P)
(a) Binomial Distribution Then (X+Y) ~ (n1 + n2 , P)
(b) Poisson distribution.

Applications of Binomial Distribution

Important continuous probability distribution Binomial distribution is applicable when the trials are independent
and each trial has just two outcomes success and failure. It is
Normal Distribution applied in coin tossing experiments, sampling inspection plan,
genetic experiments and so on.

The Chartered Accountant Student November 2021 31

STATISTICS
Poisson Distribution (vii) Additive property of Poisson distribution
Poisson distribution is a theoretical discrete probability distribution If X and y are two independent variables following Poisson
which can describe many processes. Simon Denis Poisson of France distribution with parameters m1 and m2 respectively, then Z =
introduced this distribution way back in the year 1837. X + Y also follows Poisson distribution with parameter (m1 +
m2).
Poisson Model
i.e. if X ~ P (m1)
Let us think of a random experiment under the following
and Y ~ P (m2)
conditions:
and X and Y are independent, then
I. The probability of finding success in a very small time
Z = X + Y ~ P (m1 + m2)
interval ( t, t + dt ) is kt, where k (>0) is a constant.
II. The probability of having more than one success in this
time interval is very low.
III. The probability of having success in this time interval is Application of Poisson distribution
independent of t as well as earlier successes. Poisson distribution is applied when the total number of events is
pretty large but the probability of occurrence is very small. Thus we
The above model is known as Poisson Model. The probability of
can apply Poisson distribution, rather profitably, for the following
getting x successes in a relatively long time interval T containing m
cases:
small time intervals t i.e. T = mt. is given by
a) The distribution of the no. of printing mistakes per page of a
for x = 0, 1, 2, ......……… large book.
b) The distribution of the no. of road accidents on a busy road per
Taking kT = m, the above form is reduced to
minute.
for x = 0, 1, 2, ......…..... c) The distribution of the no. of radio-active elements per minute
in a fusion process.
d) The distribution of the no. of demands per minute for health
centre and so on.
Definition of Poisson Distribution
A random variable X is defined to follow Poisson distribution with
parameter λ, to be denoted by X ~ P (m) if the probability mass
function of x is given by Normal or Gaussian distribution
The two distributions discussed so far, namely binomial and
f (x) = P (X = x) = for x = 0, 1, 2, ... Poisson, are applicable when the random variable is discrete. In
case of a continuous random variable like height or weight, it is
= 0 otherwise impossible to distribute the total probability among different mass
Here e is a transcendental quantity with an approximate value as points because between any two unequal values, there remains an
2.71828. infinite number of values. Thus a continuous random variable is
defined in term of its probability density function f (x), provided,
Important points in connection with Poisson distribution: of course, such a function really exists, f (x) satisfies the following
(i) Since e–m = 1/em >0, whatever may be the value of m, m > 0, condition:
it follows that f (x)≥ 0 for every x. f(x) ≥ 0 for x(-∞,∞) and =
Also it can be established that ∑ f(x) = 1 i.e. f(0) + f(1) + f(2)
+....... = 1 x The most important and universally accepted continuous
(ii) Poisson distribution is known as a uniparametric probability distribution is known as normal distribution. Though
distribution as it is characterised by only one parameter m. many mathematicians like De-Moivre, Laplace etc. contributed
(iii) The mean of Poisson distribution is given by m i,e µ = m towards the development of normal distribution, Karl Gauss was
(iv) The variance of Poisson distribution is given by σ2 = m instrumental for deriving normal distribution and as such normal
(v) Like binomial distribution, Poisson distribution could be distribution is also referred to as Gaussian Distribution.
also unimodal or bimodal depending upon the value of the A continuous random variable x is defined to follow normal
parameter m. distribution with parameters µ and σ2, to be denoted by
We have µ0 = The largest integer contained in m if m is a non- X ~ N (µ, σ2)
integer If the probability density function of the random variable x is given
= m and m–1 if m is an integer by
(vi) Poisson approximation to Binomial distribution f(x) = for -∞<x<∞
If n, the number of independent trials of a binomial
distribution, tends to infinity and p, the probability of a success, where µ and σ are constants, and > 0
tends to zero, so that m = np remains finite, then a binomial
distribution with parameters n and p can be approximated by a To be continued in the next capsule 
Poisson distribution with parameter m (= np).
In other words when n is rather large and p is rather small so
that m = np is moderate then β(n, p) ≅ P (m)

"Don't judge each day by the harvest you reap but by the seeds that you plant."
- Robert Louis Stevenson

32 November 2021 The Chartered Accountant Student