4) For the data in AED _HOUSE

Subject:EconomicsPrice: Bought3

Share With

4) For the data in AED _HOUSE.DTA. (a) Create three separate dummy variables d1, d2, d3 for lot size : 1, 2 and 3. (b) Regress price on size and lotsize . (c) Regress price on size , d2 and d3. Are d2 and d3 jointly statistically signi?cant (d) "Which model do you prefer , that in (b) or that in (c)? Explain . (e) Redo part (c) with d1 and d2 rather than d2 and d3. How have your answers changed? (f) Regress price on size , d1 , d2 and lotsize .Explain what has happened 5. We estimate the model yi 2,5 1 +f3 211:" +,B 3.1:" + Hi. What problems arise when ,in turn ,each of the following occurs : (a) 3:3 is an indicator variable that takes only values 0 or 1. (b) m3 : 2.132. (c) :33 should not appear in the model. (d) 354 should also have appeared in the model . (e) ui is heteroskedastic. (f) ui is correlated with 3:2. (g) "-5 is correlated with u?- . 6. The following page has Table II from Alan B. Krueger, "How Computers have Changed the Wage Structure : Evidence from N?croData , 1934 4989 ," Quarterly J. of Economics , 1993 , 33-60. You do not need the complete article, but if you want to download it then from a U.C.Davis IP address go to http :/ /www.jstor.org / stable / 2118494 and choose View PDF . The basic model is: In 115:5 1 +? 212'. + +? k1" + ac'. + as; where in ms is natural logarithm of wage and Ci is a dummy variable equal to one if the worker uses the computer at work and equal to zero otherwise We focus on the October 1989 results in Table II (columns 4 to 6). W hen determining the number of parameters in each model , also look at the footnote to the table . (a) Give the formula for the equation estimated in column (4) of table II. (b) It is claimed that computer use is associated with higher wages .Perform an appropriate one -sided test at or : .05 of the statistical signi?cance of the computer dummy variable in model (4) in table II. (c) What is the size of the estimated difference between workers who use computers on the job and those who do not given the estimate in column (4) of table 11. (d) Iscomputer use associated with higher wages after controlling for individual characteristics ?Perform an appropriate one -sided test at a : .05 of the statistical significance of the computer dummy variable in model (6) in table 11. (e) What is the effect of education on wage given the estimates in column (6) of table 11. (f) Perform an F test at a 2 .05 of the overall goodness of ?t of model (6) in table II. (g) According to the estimates in model (6) in table II, what is the difference between wages with and without use of computer , controlling for other regressors that might determine wages 7 (h) Which model do you prefer on the basis of adjusted R", model (5) or model (6)? (i) According to the interviewers ' instructions : IUsing a computer ' refers only to respondent's 'DIRECT ' or 'HANDS ON ' use of a computer with typewriter like keyboards . The computer may be a personal computer , minicomputer or mainframe computer . Would workers using electronic cash registers at MacDonalds be counted as using computers at work '2' (j) Do the results in Table II prove that using a computer at work causes higher wages? Explain.