Textbook problems:
7.4, 7.12 a,b,c,d

Computer exercises:

1. Go back to your regression on cars or scores datasets you performed in problem sets 6 and 7. What other variables might influence your dependent variable (minprice or score)? Estimate a multiple regression model with more than one explanatory variable. Is the coefficient on your original explanatory variable different? Is its interpretation different?

2. Suppose you are working for the federal government and are required to do a study of  fatalities due to highway accidents in America. Specifically, you are required to develop a regression model to explain the determinants of the number of  fatalities per licensed driver. The data set highway3.dta contains state-wise information on fatalities and potential determinants. Save the highway3.dta on a diskette or on your desktop and open it in stataquest using commands "File" and "Open." Recall that to get details regarding the data set you can choose “Summaries”, “Data Set Info” and “Describe Variables.” Do the following excercises:

(a) Write down your model and justify the inclusion of each explanatory variable. Explain what sign you would expect each coefficient to have.

(b) Present your results using the format of table 7-7 on page 225 of the textbook. In addition, put the mean of the dependent variable in the table. If  a coefficient is statistically significant at 5% (two-tail), put an asterisk against it. Include a note at the bottom of the table telling the reader what the asterisk means.

(c)  Discuss each coefficient obtained, focussing on both statistical significance and substantive significance (is it big enough to matter?).  If  appropriate, translate the coefficient into an elasticity (slope times (X*/Y*), where X* and Y* are sample means).

(d) Interpret the R2

(e)  Does your analysis have any policy implications?