You are on page 1of 21

Statistics & Research Methodology (103)

Contributors: Prof. (Gp.Capt.) D.P.Apte Prof. (Gp.Capt.) Suhas Jagdale Prof. S.N.Parasniss Prof. Anagha Gupte Prof. Anjali Mote

Session Plan Subject: Statistics & Research Methodology (103) Lecture Topic No. 1 Statistics: Introduction to business statistics, Importance, Definition & classification 2 Collection of data: Primary & secondary data, Sources and Classification of data, Tabular presentation 3 Presentation of data: Frequency distribution tables, Graphical presentation 4 Measures of Central Tendency and Variability: Ungrouped data 5 Case-1 6 Measures of Central Tendency, Measures of variability : Grouped data Fundamentals of Probability, Laws of probability Case -2 Probability: Conditional probability, Bayes theorem Probability distributions: Binomial distribution Case -3 Probability distributions: Poisson and Exponential distribution Probability distribution: Normal distribution Case - 4 Sampling : Importance and sampling techniques Sampling distribution: Concept, sampling distribution of mean Case - 5 Suggested reading a. Chapter 1 b. Chapter 1 a. Chapter 2 b. Chapter 1

a. Chapter 2 b. Chapter 1 a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. Chapter 3, 4 Chapter 1 Chapters 1, 2, 3, 7. Chapter 1 Chapter 3, 4. Chapter 1. Chapter 5. Chapter 2 Chapter 1, 2, 3, 7 Chapter 1. Chapter 5 Chapter 2 Chapter 6 Chapter 3 Chapter 5, 6 Chapters 2 Chapter 6 Chapter 3 Chapter 6 Chapter 3,4 Chapter 5, 6 Chapter 3 Chapter 7 Chapter 5 Chapter 7 Chapter 5 Chapters 5, 6

7 8 9 10 11 12 13 14 15 16 17

18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Sampling distribution of proportion and variance Estimation for decision making Case - 6 Concepts of Hypothesis Testing Tests of significance - I Case - 7 Tests of significance - II Tests of significance - III Case-8 Test of significance - IV Correlation: Simple correlation, Multiple and partial correlation Case - 9 Regression analysis I: Simple correlation Regression analysis II: Multiple correlation Case - 10 Analysis of variance Surprise Test - 1 Surprise Test- 2

b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b. a. b.

Chapters 4 Chapters 7 Chapters 5, 6 Chapter 6,7, 8 Chapter 5, 6 Chapter 6, 7 Chapters 4, 5 Chapter 9 Chapter 7 Chapter 9, 10, 11 Chapter 7, 8 Chapters 9, 10 Chapters 7, 8 Chapters 9, 10, 11 Chapters 7, 8 Chapters 10, 11 Chapters 7, 8 Chapter 9, 10,11 Chapter 8 Chapter 9, 10, 11 Chapter 7, 8 Chapter 12 Chapter 10 Chapter 14 Chapter 9 Chapter 13 Chapter 10 Chapter 13 Chapter 11 Chapter 12, 13 Chapter 10 Chapter 14 Chapter 9

Suggested Reading a. Statistical tools for Managers: Using MS Excel By Prof. (Gp.Capt.) D.P.Apte. b. Complete Business Statistics: By Amir D. Aczel; Jayavel Sounderpandian.

Data for case 1 : Better to be roughly right than precisely wrong

S N

SSC %

HSC %

S N

SSC %

HSC %

S N

SSC %

HSC %

SN

SSC %

HSC %

SN

SSC %

HSC %

SN

SSC %

HSC %

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

77.29 75.73 62.66 71.06 71.40 67.00 62.00 61.40 83.71 78.20 54.00 50.26 73.40 57.80 44.00 88.32 64.80 68.00 71.20 55.20 81.80 69.46 71.20 70.80 55.33 75.46 55.20 62.80 70.00

65.57 73.17 74.00 68.50 66.00 69.00 56.00 65.80 74.00 78.60 48.50 62.67 68.40 50.33 51.11 83.16 48.50 69.40 68.60 65.40 75.40 56.50 81.50 61.80 53.83 57.33 61.50 65.20 71.57

30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58

75.60 44.26 59.20 60.00 62.16 61.33 67.60 80.66 66.53 76.80 71.16 76.66 74.80 62.71 64.80 60.40 76.60 75.60 79.00 69.90 71.60 63.73 77.50 60.00 75.20 80.86 61.40 59.20 80.40

75.33 45.00 53.00 62.83 59.33 75.33 55.60 76.33 61.67 61.40 61.80 54.67 52.83 56.14 70.83 51.00 68.20 49.83 72.80 75.00 69.67 62.17 67.20 61.00 59.40 56.00 69.40 49.33 63.50

59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87

66.26 56.00 74.66 57.20 59.66 59.50 51.20 75.80 59.46 69.46 70.13 80.80 72.00 60.00 59.33 62.00 51.40 59.00 72.93 60.00 84.16 59.66 72.80 77.86 59.20 57.00 64.26 51.60 56.80

55.33 57.14 60.67 56.88 61.33 50.33 76.40 73.60 62.00 69.67 49.00 75.25 77.00 59.33 70.50 71.20 57.80 56.20 51.50 72.00 64.90 68.20 68.60 73.00 75.83 57.00 61.00 47.40 77.00

88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116

67.66 82.00 63.73 71.06 60.43 57.80 87.33 72.66 58.00 80.36 68.14 55.14 55.86 48.66 61.00 74.26 58.00 67.80 85.06 70.66 62.26 64.80 70.00 53.20 56.40 61.20 52.80 72.40 54.40

61.00 66.80 49.83 49.67 71.29 61.40 78.00 66.60 40.60 80.75 76.29 49.50 50.67 55.17 61.66 75.00 59.80 63.83 55.83 72.33 62.67 71.67 69.33 52.20 65.33 67.55 60.83 65.30 56.33

117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145

53.80 72.00 47.40 44.66 64.00 74.33 76.20 73.80 70.20 56.00 68.00 70.53 54.00 62.80 49.86 64.13 65.00 56.40 61.86 71.80 74.40 47.20 77.00 60.40 63.60 56.40 62.83 63.66 57.60

70.33 64.83 60.00 61.80 49.80 60.80 83.20 63.20 59.60 58.83 57.33 66.67 74.40 65.20 65.83 61.00 58.22 56.20 50.83 72.16 74.60 71.83 82.60 48.67 59.00 48.66 62.40 61.33 49.00

Subject: Statistics & Research Methodology (103) Case No: 02

Note: Sample questionnaire is attached. However, please feel free to add few question based on your purpose.

MIT School of Business Students Profile

Batch Roll No

Specialization for PGDM ______________________ Name ______________________________________________________ Date of Birth (MM/DD/YY) -----/-----/------ Blood group Email address:_________________________ Contact no.: Landline: _________________ Mobile: _________________ Present Address:_____________________________________________ Permanent Address:___________________________________________ Educational Qualification i) ii) iii) iv) 10th percentage _______% 12th percentage _______% Graduation Percentage ________% Graduation Type: Arts / Commerce / Science / Engineering / Management / Pharmacy/ Other (Specify)___________________ v) vi) vii) viii) Name of the College (for graduation):________________________________ College Address:________________________________________________ Work experience: Yes / No Whether attended coaching for competitive exams (CAT/MAT/CET etc) Yes / No If Yes, mention the name of the institute:_____________________________ Payment of the current fees by i) Own ii) Loan Future plans: i) Job ii) Family Business iii) Own business iv) First job then business Family Background Fathers Education: _______________ Fathers occupation:___________________ Mothers Education: _______________ Mothers occupation:___________________ Family Type: Nuclear / Joint / Single parent _________

Number of family members staying together:___________ Number of rooms in the house:______ Family income (in Rs. Per year):_____________ Newspaper read: __________________________ Popular radio channel in your city:_________________ How did you come to know about MITSOB i) ii) iii) Newspaper Advertise (Specify)____________________ Internet : MITSOB website / Google / Facebook / Shiksha / Other Recommendation from friends

If you want to recommend MITSOB for your junior friends from college, give contact no. of such friends. i________________ ii __________________ iii____________________

------------------XXXXXX--------------

Subject: Statistics & Research Methodology (103) Case No: 03 Job Applications

A business graduate very much wants to get a job in any one of the top 10 accounting firms. Applying to any one of these companies requires at lot of effort and paperwork and is therefore costly. She estimates the cost of applying to each of the 10 companies and the probability of getting a job offer there. These data are tabulated below. The tabulation is in the decreasing order of cost. 1. If the graduate applies to all 10 companies, what is the probability that she will get at least one offer? 2. If she can apply to only one company, based on cost and success probability criteria alone, should she apply to company 5? Why or why not"? 3. If she applies to companies 2, 5, 8, and 9, what is the total cost? What is the probability that she will get at least one offer? 4. If she wants to be at least one 75% confident of getting at least one offer, to which companies should she apply to minimize the total cost? (This is a trial-and-error problem.) 5. If she is willing to spend \$1,500, to which companies should she apply to maximize her chances of getting at least one job? (This is a trial-and-error problem.) Company Cost 1 2 3 4 5 6 7 8 9 10

\$870 \$600 \$540 \$500 \$400 \$320 \$300 \$230 \$200 \$170 0.38 035 0.28 0.20 0.18 0.18 0.17 0.14 0.14 0.08

Probability

* Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.

Subject: Statistics & Research Methodology (103) Case No: 04 Microchip Contract A company receives an order for five custom-made microchips at a price of \$7,500 each. The company will produce the chips one by one using a complex process which has only a 67% chance of producing a defect-free chip at each trial. After five defect-free chips are produced the process will be stopped. A cost accountant at the company has prepared the following cost report: The cost of production includes a \$14,800 fixed cost and a \$2,700X unit variable cost. Thus if X number of chips are produced, the total cost of production would be 14,800 + 2.700X dollars. The revenue minus the cost of production will be the profit. After some analysis the finance manager of the company says that the risk may be too high and thinks the order should not be accepted. 1. What distribution will the number of chips produced, X, follow? 2. What is the expected value and standard deviation of X? 3. What is the expected value and standard deviation of the profit? 4 What is the break-even X (allow fractional values for X)? 5. What is the probability that accepting the order will result in a loss? 6. A popular measure of risk in a venture is value at risk, which is the loss suffered at the 5th percentile of the return from the venture. In this problem, find an integer x such that P[X >x] is approximately 5%. 7. For the x value found in part 6, calculate the loss, and thus the value at risk. 8. Express the value at risk as a percentage of the expected value of the profit. 9. What is your assessment of the risk and reward in the order? Should the company accept the order? The sales manager of the company says that the customer is very likely to agree to increase the order quantity from five to eight chips. But he is not sure whether the matter should be pursued with the customer. 10. "If accepting an order of five itself is risky, will it not be even more risky to accept an order for eight?" asks the sales manager. How would you answer him?

11. Calculate the expected value and standard deviation of the profit for an order quantity of eight. 12. What is the value at risk for an order quantity of eight, computed in a manner similar to parts 6 and 7 above? Express the value at risk as a percentage of the expected profit. 13. Looking at the answer to parts 3,8,11, and 12, would you say the risk and reward have become more favorable, compared to an order quantity of five? 14. Should the company pursue the matter of increasing the order quantity to eight with the customer? * Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.

Case No: 07 Tiresome Tire I When a tire is constructed of more than one ply, the interply shear strength is an important property to check. The specification for a particular type of tire calls for a strength of 2,800 pounds per square inch (psi). The tire manufacturer tests the tires using the null hypothesis H0 : u> 2,800 psi Where u is the mean strength of a large batch of tires from past experience, it is known that the population standard deviation is 10 psi. Testing the shear strength requires a costly destructive test and therefore the sample size needs to be kept at a minimum. A type I error will result in the rejection of a large number of good tires and is therefore costly. A type II error of passing a faulty batch of tires can result in fatal accidents on the roads, and therefore is extremely costly. (For purposes of this case, the probability of type II error, 3, is always calculated at u =2,790 psi.) It is believed that should be at most 1%. Currently, the company conducts the test with a sample size of 40 and an a of 5% To help the manufacturer get a clear picture of type I and type II error probabilities, draw a (3 versus a chart for sample sizes of 30, 40, 60, and 80. If 3 is to be at most 1 % with = 5% which sample size among these four values is suitable? 1. Calculate the exact sample size required for = 5% and 3 = 1%. Construct a sensitivity analysis table for the required sample size for u ranging from 2,788 to 2,794 psi and 3 ranging from 1% to 5%. 2. For the current practice of n = 40 and = 5% plot the power curve of the test. Can this chart be used to convince the manufacturer about the high probability of passing batches that have strength of less than 2,800 psi? 3. To present the manufacturer with a comparison of a sample size of 80 versus 40, plot the OC curve for those two sample sizes. Keep an of 5%. 4. The manufacturer is hesitant to increase the sample size beyond 40 due to the concomitant increase in testing costs and, more important, due to the increased time required for the tests. The production process needs to wait until the tests are completed, and that means loss of production time. A suggestion is made by the production manager to increase to 10% as a means of reducing . Give an account of the benefits and the drawbacks of that move. Provide supporting numerical results wherever possible. * Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.

Subject: Statistics & Research Methodology (103)

Case No: 08 Tiresome Tire II A tire manufacturing company invests a new, cheaper method for carrying out one of the steps in the manufacturing process. The company wants to test the new method before adopting it, because the method could alter the interply shear strength of the tires produced. To test the acceptability of the new method, the company formulates the null and alternative hypothesis as H0 :1 - 2 < = 0 H1:1 -2 >0 Where 1 is the population mean of the interply shear strength of the tires produced by the old method and 2that of the tires produced by the new method. The evidence is gathered through a destructive test of 40 randomly selected tires from each method. Following are the data No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Sample 1 2792 2755 2745 2731 2799 2793 2705 2729 2747 2725 2515 2782 2718 2719 2751 2755 2685 2700 2712 2778 Sample 2 2713 2741 2701 2731 2747 2679 2773 2676 2677 2721 2742 2775 2680 2786 2732 2740 2760 2748 2660 2789 No. 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Sample1 2693 2740 2731 2707 2754 2690 2797 2761 2760 2777 2774 2713 2741 2789 2723 2713 2781 2706 2776 2738 Sample2 2683 2664 2757 2736 2741 2767 2751 2723 2763 2750 2686 2727 2757 2788 2676 2779 2676 2690 2764 2720

1. Test the null hypothesis at a = 0.5.

Later it was found that quite a few tires failed on the road. As a part of the investigation, the above hypothesis test is reviewed. Considering the high cost of type II error, the value of 5% for a is questioned. The response was that the cost of type I error is also high because the new method could save millions of dollars. What value for a would you say is appropriate? Will the null hypothesis be rejected at that a? 2. A review of the tests conducted on the samples reveals that 40 otherwise identical pairs of tires were randomly selected and used. The two tires in each pair underwent the two different methods, and all other steps in the manufacturing process were identically carried out on the two tires. By virtue of this fact, it is argued that a paired difference test is more appropriate. Conduct a paired difference test at a = 0.5. 3. There is a move to reduce the variance of the strength by improving the process. Will the reduction in the variance of the process increase or decrease the chances of type I and type I errors? * Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.

Subject: Statistics & Research Methodology (103) Case No: 09 Uniform Uniforms
A textile manufacturer has a large order for a cloth meant for making uniforms. The cloth is dyed using four different dyeing lines, which produce approximately equal amounts of cloth each day. Usually not more than one line is used for one product, because no matter how well the process is controlled, there will always be perceptible differences in the shade of the dye from one line to another. But because the volume of the order is large, four lines are being used. It is important to maintain the shade as uniform as possible by minimizing the variance of the brightness of the shade on all cloth produced. Lately, the customer has been complaining about too much variance in the brightness. It was decided to conduct an ANOVA test of the brightness of the cloth from the four lines. Random samples were taken from each line and were measured for brightness. The measurement is on a 0 to 100 scale. The sample data are Line1 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 66.55 71.91 67.61 66.13 71.31 68.99 71.83 68.99 69.81 72.49 69.99 73.44 70.39 68.42 71.66 65.14 Line2 66.16 65.94 68.62 63.86 69.38 64.55 66.82 65.56 63.66 64.71 67.32 71.39 63.78 70.42 Line3 68.36 66.81 66.50 65.22 65.06 65.42 66.50 64.82 68.31 68.17 65.50 70.39 Line4 72.32 66.69 72.36 70.88 71.05 71.05 68.78 74.40 73.58 73.58 66.72 70.37 75.72 74.65

1. Conduct the test at the 5% significant level, and report your conclusion. 2. Which pairs of lines have significant differences in their average brightness? 3. Stopping a line to adjust its average brightness is costly. If only one line can be stopped and adjusted, which one should it be? To what average brightness value should it be adjusted to minimize the variance in all the cloth produced? 4. If two lines can be stopped and adjusted, which ones should be? To what average brightness value should they be adjusted to minimize the total variance in all the cloth produced? * Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.

Subject: Statistics & Research Methodology (103) Case No: 10 Risk and Return According to the Capital Asset Pricing Model (CAPM) the risk associated with a capital asset is proportional to the slope b obtained by regressing the assets past returns with the corresponding returns of the average portfolio called the market portfolio . ( The return of the market portfolio represents the return earned by the average investor. It is weighted average of the returns from all the assets in the market.) The larger the slope b of an asset , the larger is the risk associated with that asset. A b of 1 represents average risk. The Returns from an electronic Firms stock and the corresponding returns from the market portfolio for the past 15 years are given below.

Market return (%) 16.02 12.17 11.48 17.62 20.01 14 13.22 17.79 15.46 8.09 11 18.52 14.05 8.79 11.6

Stocks return (%) 21.05 17.25 13.1 18.23 21.52 13.26 15.84 22.18 16.26 5.64 10.55 17.86 12.75 9.13 13.87

Carry out the regression and find the B for the stock. What is the regression equation? 1. Does the value of the slope indicates that the stock has above-average risk? (For the purposes of this case assume that the risk is average if the slope is in the range 1 0.1, below average if it is less than 0.9, and above average if it is more than 1.1.) 2. Give a 95% confidence interval for this 3 Can we say the risk is above average with 95% confidence?

3. If the market portfolio return for the current year is 10%, what is the stock's return predicted by the regression equation? Give a 95% confidence interval for this prediction. 4. Construct a residual plot. Do the residuals appear random? 5. Construct a normal probability plot. Do the residuals appear to be normally distributed? 6. (Optional) The risk-free rate of return is the rate associated with an investment that has no risk at all, such as lending money to the government. Assume that for the current year the risk-free rate is 6%. According to the CAPM, when the return from the market portfolio is equal to the risk-free rate, the return from every asset must also be equal to the risk-free rate. In other words, if the market portfolio return is 6%, then the sock's return should also be 6%. It implies that the regression line must pass through the point (6,6). Repeat the regression forcing this constraint. Comment on the risk based on the new regression equation. * Case taken from Complete Business Statistics by Amir Aczel & J. Sounderpandian, TATA Mc Graw Hill.