You are on page 1of 2

Faculty of Management Studies, University of Delhi

MBA (PT) I yr, Semester I, 2016-2017 8102: Quantitative Methods for Management
Assignment – V – Sampling
Submitted By:
Shakher saini , Roll no: S070 , South Campus
Question 1
One of the major oil distribution companies conducted a study recently to estimate the mean quantity of high
speed petrol purchased by customers of a particular brand of car per visit to a petrol pump station. To do this,
a random sample of customers was selected with the following data being recorded that show the liters of
high speed petrol purchased by owners of that particular brand of car.
(a) Based on these sample data, construct and interpret a 95 percent confidence interval estimate for the
population mean.
Ans 1.(a) From the given data the we can calculate the descriptive statistics as mentioned below :

Actual Data (n=30)   Actual Data


13.3 24    
15.1 24.2 Mean of the sample (x ̅ ) 23.55
15.7 24.4 Standard Error (s/√n) 0.9837
15.9 24.4 Median 23.7
18.1 24.9 Mode 24.4
18.7 25.7 Standard Deviation () 5.3877
18.9 25.7 Sample Variance 29.0274
19.9 27.1 Kurtosis -0.1362
21.2 28.4 Skewness 0.1857
21.3 28.9 Range 21.7
22 29 Minimum 13.3
22.3 29.5 Maximum 35
22.4 30 Sum 706.6
22.5 34.7 Count (n) 30
23.4 35   Confidence Level (95.0%) 2.012

Apart from using excel Standard error of the mean can also be calculated by its formula,
Standard Error of the mean = (s/√n) = (5.3877/√30) = 0.9837

As degrees of freedom approach infinity, the t distribution approaches the z distribution so the value of


Zα/2 from the standard table , 2.5% on both sides gives , 95%+2.5% on each side = 1.96

Using the formula for confidence interval : x̅ ± Zα/2(s/√n)

95% C.I : 23.55 ± 1.96(.9837) = 23.55 ± 1.928 = [Lower level is 21.62 ,Upper level is 25.48]
So we are 95% confident that the mean quantity of high speed petrol purchased by customers of a
particular brand of car is between 21.62 & 25.48 liters.

1.(b) A manager in that oil company is interested in estimating the mean herself through another sample
survey and wants to develop a confidence interval estimate. Suppose, she has determined the required
sample size and feels that she cannot afford one that large. What options are available?

Ans 1.(b) No idea sir , needs to learn in class this one ,


I guess Using point estimators or using (s) as an estimate of (σ) can be used if n>30 but that would add extra
variability; so it would not be appropriate to use the normal curve multiplier associated with 95% confidence.
Question 2
A major tire manufacturer wishes to estimate the mean tread life in miles for one of its tires. It wishes to
develop a confidence interval estimate that would have a maximum sampling error of 500 miles with 90
percent confidence. A pilot sample of n = 50 tires showed a sample standard deviation equal to 4,000 miles.
Based on this information, what is the required sample size?

Ans 2 : The following data is given in the question ,


Max Sampling Error = 500 miles with confidence interval of 90% corresponding value of which in the table
comes out to be 1.645 to be used in the calculation.
n = 50 tires & standard deviation (σ)= 4000 miles.
N=?
Using the formula for Sampling error i.e S.E = 1.645(σ /√N) at 90% confidence interval.

Putting the values , we get N = {(1.645*4000)/500}*{(1.645*4000)/500}

SO the required sample size would be 174 tires approximately.

You might also like