The variables are self-explanatory except perhaps for AGE, the average age of the adult members of the household. The data are in the file vacation.dat.

(a) The equation was estimated by least squares and the residuals are plotted against age and income in Figure 8.4. What do these graphs suggest to you?

(b) Ordering the observations according to descending values of INCOME, and applying least squares to the first 100 observations, and again to the second 100 observations, yields the sums of squared errors SSE1 2:9471107 SSE2 1:0479107 Use the Goldfeld-Quandt test to test for heteroskedastic errors. Include specification of the null and alternative hypotheses.

(c) Table 8.2 contains three sets of estimates: those from least squares, those from least squares with White's standard errors, and those from generalized least squares under the assumption s2 i s2 INCOME2.

(i) How do vacation miles traveled depend on income, age, and the number of kids in the household?

(ii) How do White's standard errors compare with the least squares standard errors? Do they change your assessment of the precision of estimation?

(iii) Is there evidence to suggest the generalized least squares estimates are better estimates?