Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

Principles of Econometrics ECMT 5001

Take-home Mid-semester Test

2020

1. The general manager of the Hilton Hotel in Sydney is evaluating an employment screening test for the front office clerical staff. During this evaluation all new clerical employees are given the test.

70% pass the test; the rest fail. At a later time, after the new clerical employees have been working for a while, their performance is evaluated as being satisfactory or unsatisfactory. Historically, 80% of all clerical employees have been found to be satisfactory, and 75% of the satisfactory clerical employees in the evaluation of the employment screening test have passed the screening test.

(a) From the given information, determine the probabilities of the following events:  (i) pass- ing the test, (ii) having satisfactory performance and (iii) passing the test given satisfactory performance.                                                                                                                     [3 marks]

(b) Using your answers to part (a), determine the probability of a clerical employee passing the test and having satisfactory performance.                                                                        [1 mark] (c) Using your answer in parts (a) and (b), determine the probabilities of the a clerical employee: (i) failing the test and having satisfactory performance, (ii) failing the test and having unsat- isfactory performance, (iii) passing the test and having unsatisfactory performance, and (iv) having unsatisfactory performance.                                                                                [4 marks]

(d) Using your answers in part (c), determine the probabilities of a clerical employee: (i) failing the test given they are found to have unsatisfactory performance, (ii) failing the test given they are found to have satisfactory performance, (iii) passing the test given they are found to have unsatisfactory performance, (iv) unsatisfactory performance given they failed the test, (v) satisfactory performance given they passed the test, (vi) unsatisfactory performance given they passed the test, and (vii) satisfactory performance given they failed the test.     [7 marks] (e) Using your answers in part (d), determine the following percentages: (i) clerical employees who failed the test and prove to be unsatisfactory, and (ii) clerical employees who passed the test and who prove to be satisfactory.                                                                             [2 marks]

(f)  Government guidelines require screening tests to achieve at least 20% for part (i) in (e) and

at least 60% for part (ii) in (e). Does this test meet those government requirements? Explain. [2 marks]

2. In designing the power system for a satellite, an engineer needs to determine the probability of early failure for the batteries powering the system.  A random sample of batteries needs to be selected for testing for early failure. Those sampled batteries will all be of the same type, suitable for use in the satellite.  The engineer wishes to find the sample size that will provide an estimate of the true proportion of batteries that do not fail too early, 工, that will be within 0 ·05 of the population proportion with probability no less than 95%. Assume that the population of batteries is large.

(a) If 工 is initially assumed to be 0.5, find the engineer’s required sample size, explaining your

reasoning.                                                                                                                         [5 marks]

(b) After testing the number of batteries found in (a), only 250 prove satisfactory.  What is the point estimate of 工? If this were the true value of 工 how many batteries were tested unneces- sarily?                                                                                                                                [5 marks]


3. Your friend owns a retail store where she plans to give arriving customers a short verbal ”com- mercial” message regarding products for sale. Your friend asks you to determine if this will boost the average dollar value of sales to each customer, µ, which is currently $5.00. You may assume that the standard deviation per transaction is $1.00 and that the population is of unlimited size. You monitor a random sample of n = 200 customers given the short verbal ”commercial” message upon arrival at the store and determine the amount purchased by each.

(a) Your friend will only adopt the policy of giving short verbal ”commercial” messages to each newly arriving customer if the sample indicates that the values of sales per customer will improve. Formulate the null and alternative hypotheses.                                              [2 marks]

(b)  Suppose you wish to test the null hypothesis while ensuring that your friend only has a 5% chance of getting a recommendation to adopt the policy of giving short verbal ”commercial” messages when that policy actually does not improve the value of sales per customer. Describe the test statistic you would use to conduct the test, compute the critical region for the test, and describe the decision rule you would use.                                                                      [4 marks]

(c) You collect the sample data.  What recommendation would you make to your friend if (i) X¯=$5.37, (ii) X¯=$5.05, (iii) X¯=$4.97, and (iv) X¯=$5.20 ? Express your recommendations as


formal hypothesis testing conclusions.


[4 marks]