MODULE IV THEORETICAL DISTRIBUTION
(PROBABILITY DISTRIBUTION)
Basic assumptions and characteristics-Probability distribution-Fitting of
probability distribution-Binomial distribution-Fitting of binomial
distribution-Poisson distribution-Fitting of Poisson distribution-Normal
distribution-Features and properties-Standard normal curve.
By theoretical distribution, we mean a frequency distribution, obtained for a
random variable, on the basis of a mathematical model. Binomial distribution,
Poisson Distribution and Normal Distribution are examples of theoretical
distributions.
A theoretical distribution in statistics assumes that a set of data follows a
specific pattern with defined characteristics like a fixed number of trials,
independent events, and a constant probability of success for each trial, allowing
for calculations of probability based on these assumptions, with common examples
including the Binomial, Poisson and Normal distributions.
Assumptions and Characteristics: Important assumptions are:
Fixed number of trials: The total number of experiments or observations is
predetermined and remains constant throughout the analysis.
Independent events: Each trial or observation is independent of the others,
meaning the outcome of one event does not influence the the outcome of
another.
Constant probability of success: The probability of a success event remains
the same for each trial.
The main characteristics of theoretical distributions are:
The shape of the distribution curve
Mean
Variance
Standard Deviation
Probability distribution is a description which gives the probability
for each value of the random variable which is often expressed in the format
of a graph, table or formula. It describes what will probably happen instead
of what happened.
Fitting of probability distribution:
Fitting a probability distribution refers to the process of selecting a
theoretical probability distribution that best matches the pattern of a given
set of data, essentially finding the distribution that most accurately describes
the likelihood of different outcomes within the data set, by analyzing its
probability density function.
Steps involved in fitting a probability distributions are:
Visualize the data
Select potential distributions
Parameter estimation
Goodness-of-fit tests.
Types of theoretical distributions: Theoretical distributions are of many
types. The most popular and widely used theoretical distributions are:
A. Binomial Distribution
It is the most important and widely used discrete probability distribution. It
can be understood as the probability of a trial with two and only two
outcomes as “success” and “Failure”. It is applicable to discrete random
variables only.
Requirements of a Binomial Experiment: A binomial experiment is a
statistical experiment which meets all the following requirements:
The experiments consists of finite number (n) of repeated trials.
The outcome of each trial should have exactly two categories, commonly
referred to as either success and failure [(p) or (q)].
The values of both p and q will be between 0 to 1 and the sum of these
values (p and q) will be always 1.
The probability of success (denoted by p) remains the same in all the trials.
The trials are independent of each other. This means that the outcome of one
trial does not affect the outcome on other trials.
Situations where Binomial distributions can be applied:
The random experiment has two outcomes, which can be called “success”
and “failure”.
Probability for success in a single trial remains constant from trial to trial of
the experiment.
The experiment is repeated finite number of times.
Trials are independent.
Properties or Characteristics of Binomial Distribution:
Binomial distribution is a discrete probability distribution.
The shape and location of Binomial Distribution changes as ‘p’ changes for
a given ‘n’.
Binomial distribution has one or two modal values.
Mean of the Binomial Distribution increases as ‘n’ increases, with ‘p’
remaining constant.
If ‘n’ is large and if either ‘p’ nor ‘q’ is too close to zero, Binomial
Distribution has mean= np and S.D=√ npq .
If two independent random variables follow Binomial Distribution, their
sum also follows Binomial Distribution.
Importance of Binomial distribution:
The binomial distribution is often very useful in decision making situations
in business.
In quality control it is very widely applied.
The Binomial Distribution is used in sampling techniques.
The Binomial Distribution is used in such sampling techniques.
In accepting sampling plans, inspections is carried out on the items drawn in
a sample.
Constants of Binomial distribution:
Mean= np
Standard deviation =√ npq
Variance= npq
The first moment of the mean µ1=0
The second moment of the mean µ2=npq
The third moment of the mean µ3= npq(q-p)
Binomial formula: p(x)= nCxpxqn-x
Fitting a Binomial Distribution:
Determine the values of p and q and n and substitute them in the function
nCxpxqn-x, we get the probability function of the Binomial Distribution.
Put x= 0,1,2,…… in the function nCxpxqn-x, we get n+1 terms.
Multiply each such term by N (total frequency), to obtain the expected
frequency.
B. Poisson Distribution
The Poisson distribution is a discrete probability distribution for the
counts of events that occur randomly in a given interval of time, space, distance
etc. Poisson distribution may be obtained as a limiting case of Binomial
distribution under the following conditions:
The number of trials is very large.
The probability of success for each trial is very small
‘np’ is finite(say ‘m’)
Poisson equation:
P(x) = e-mmx/ x! , we obtain the probabilities for 0,1,2,3,……..times the event
to happen.
If we know ‘m’, all the terms of the Poisson Distribution can be obtained.
“m” is the parameter of the Poisson Distribution. It is the mean of the
poisson distribution.
Properties or Characteristics of Poisson Distribution
Poisson distribution is a discrete probability distribution.
If ‘x’ follows a Poisson Distribution, then ‘x’ takes values 0,1,2,….to
infinity.
Poisson Distribution has a single parameter, m. When ‘m’ is known all the
terms can be found out.
Mean and variance of Poisson Distribution are equal to m.
Poisson Distribution is a positively skewed distribution.
Assumptions:
A Poisson distribution is the probability distribution from a poisson
experiment. A distribution is treated as Poisson distribution only if the
following assumptions are valid:
The average number of success that occurs in a specified interval is known.
The occurrences of the events must be random.
The experiment results in outcomes which can be classified into success or
failure and the sum of the probabilities of these outcomes is 1.
The occurrences must be independent of each other. An event should not
influence the outcome of another event.
Fitting a Poisson Distribution
Poisson distribution is Uni -parametric in [Link] in order to prepare the
probability distribution of a Poisson distribution, the only one parameter
required is the average. To fit a Poisson distribution, each probability is
multiplied with the total number (N).
C. Normal Distribution
Normal distribution is a probability function which describes how the
values of a variable are distributed. It is a continuous probability distribution. It is
symmetrical, bell shaped curve and hence, the right side of the center is a mirror
image of left side. It is the most important probability distribution in statistics
because it fits many natural phenomena such as human height, weight,
temperature, blood pressure, sugar count etc.
Parameters of Normal Distribution
The parameters of normal distribution are the mean and standard deviation. In
order to fit a normal distribution, these two values are to be known.
Properties or Characteristics of the Normal Distribution
The Normal distribution is a continuous curve.
The Normal curve is bell shaped.
It is symmetrical about the mean.
Mean, median and Mode are equal for a Normal Distribution.
The height of Normal Curve is at its maximum at the mean.
There is only one maximum point, which occurs at the mean.
The ordinate at mean divides the whole area into two equal parts (ie 0.5 on
either side).
Q1 and Q3 are equidistant from median.
Normal distribution is mesokurtic.
If X and Y are two independent Normal variates, then their sum is also a
normal variate.
Co-efficient of skewness is 0.
Standard Normal Curve
If X is a random variable following normal distribution with mean and
standard deviation, then the variable Z= x-mean/ standard deviation is
known as standard normal variate. This ‘Z’ follows normal distribution with
mean 0 and S.D =1. The distribution of Z is known as Standard Normal
Distribution.
Importance or Uses of Normal Distribution.
Most of the discrete probability distributions tend to Normal distribution as
‘n’ becomes large.
Almost all sampling distributions conform to the normal distribution for
large values of ‘n’.
Values tests of significance are based on the assumption that the parent
population from which samples are drawn follows Normal distribution.
It is extensively used in large sampling theory to find estimates of
parameters from statistics, confidence limits etc.
In theoretical statistics as well as applied works many problems can be
solved only under the assumption of a normal distribution.
The normal distribution has numerous mathematical properties which make
it popular and comparatively easy to manipulate.
Merits of Normal Distribution
Normal distribution is the mostly used distribution in inferential statistics.
Most of error of measurements and a large variety of physical observations
have approximately Normal distributions.
The measurements of linear dimensions of large number of articles produced
may show individual variations. These follow Normal distribution.
The standard Normal distribution table shows exhaustively areas for the
different intervals of the values of the variable.
The Normal distribution has a number of mathematical properties.
Demerits
The variables which are not continuous cannot be normally distributed.
Therefore many distributions in Economics cannot be studied under Normal
Distribution.
The Normal distribution cannot be applied to situations where the
distribution is highly skewed.
Fitting Normal distribution
Parameters of the Normal distribution are the mean and standard deviation.
Find mean and standard deviation of the given frequency distribution. Take
the as the estimates of the parameters mean and standard deviation. Then we
get the density function.
**************
