Scribd
Upload
61 views

Quantitative Methods I: Section C (Computer) Assignment - 2 Nikhil Vishal Bagde 1211174

The document contains two quantitative assignments involving probability distributions. [1] The first assignment involves calculating probabilities using a binomial distribution based on birth rate data and a Poisson distribution based on bad check data received by a bank. [2] The second assignment involves calculating the posterior probability distribution of a company's market share using Bayes' theorem, based on prior probability distributions and survey sample data of 25 people. The results show an 82% probability that the company's market share lies between 35-45%.

Uploaded by

Nikhil Bagde
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
61 views

Quantitative Methods I: Section C (Computer) Assignment - 2 Nikhil Vishal Bagde 1211174

The document contains two quantitative assignments involving probability distributions. [1] The first assignment involves calculating probabilities using a binomial distribution based on birth rate data and a Poisson distribution based on bad check data received by a bank. [2] The second assignment involves calculating the posterior probability distribution of a company's market share using Bayes' theorem, based on prior probability distributions and survey sample data of 25 people. The results show an 82% probability that the company's market share lies between 35-45%.

Uploaded by

Nikhil Bagde
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOC, PDF, TXT or read online on Scribd
You are on page 1/ 4

Quantitative Methods I: Section C (COMPUTER) ASSIGNMENT 2 Nikhil Vishal Bagde 1211174

Q1 i) Random variable whose distribution can be modeled as Binomial Distribution

As per the latest provisional census report 2011, the sex ratio of children ranged 0-6 years in India is 914 female per 1000 males (pib.nic.in/prs/2011/latest31mar.pdf). The distribution is Binomial as the probability of birth of child is independent of other child and is constant. Assume that no prenatal diagnostic test to determine the sex of the unborn baby is done which can affect the birth of the girl child. The population size is very large and sampling is done without replacement. Now use the probabilities from the given sex ratio data and calculate the probability when 10 newly born babies are randomly selected from the maternity division of a Government Hospital in a Bangalore on a particular day that 3 female child are born. Probability of birth of a girl child=914/(1000+914)=0.4775
P(X = 3) = 10C3 (0.4775) 3 (1-0.4775)10-3 P(X = 3) = 0.1389

ii) Random variable whose distribution is discrete but should not be modeled as

Binomial Distribution

Consider the State Bank of Mysore in the IIM-B campus. Suppose the bank receives on an average 4 bad checks everyday. The random variable here is the number of bad checks received each day which can be X=0,1,2,3,4,5,6,This distribution is a Poisson Distribution. The number of bad checks received each day is independent and also the probability of receiving very high number of bad checks is low i.e. the probability is different in each case. Hence it can not be modeled as Binomial Distribution. Now using the above data we calculate the probability that bank receives 6 bad checks on a particular day

P ( X = i ) = f x (i ) = P(X=6)=(e^-4*4^6)/6!=0.1042

e i , i = 0,1, 2, i!

Q2 a&b) This below data is the prior probability distribution data for calculating market share. We assume the possible values of market share ranges from 20% to 65%. Also it is known from prior surveys that the market share of the company lies between 35% to 65%.
Prior Probability Distribution
S (Market share of the company) 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% Total Probability P(S) (probability of the company having the market share S) 0.02 0.05 0.12 0.29 0.28 0.11 0.06 0.04 0.02 0.01 1

c) We consider a random sample of 25 people (n=25) and 10 people have responded positively for the survey (x=10).

d) After the calculations using Bayes Theorem, the posterior probabilities are as below (The Binomial Revision.xlsx available in the companion CD of book Complete Business Statistics 7e Aczel, et. al. is used to calculate the posterior probabilities)

Bayesian Revision.xls

Posterior Probability Distribution


S (Market share of the company) 20% 25% 30% 35% 40% 45% 50% 55% 60% 65% Total Probability P(S) (probability of the company having the market share S) 0.0019 0.0169 0.0892 0.3313 0.3659 0.1266 0.0474 0.0169 0.0034 0.0005 1

As evident from the posterior probabilities table, the total posterior probability of three market share values 35%, 40% and 45% is equal to 0.8238. The three adjacent values are thus a set of highest posterior probability and can be taken as credible set of values for S with posterior probability close to the 82 % confidence level. Hence from the calculations we are 82% confident that the market share lies between 35% to 45%. This validates the prior assumption that the company has market share between 35% to 45%.

You might also like