Question 6

QUESTION 6 For this question use information in column (2) of question 3. Can you test for the exogeneity of the instrument in this model? 0 a. Yes, I can use a ttest. O b' I don't need to test. instruments are always exogeneous. O c. Yes. I can use a Jtest, even if it is not reported in the table. 0 d' No. in order to test for exogeneity. the model must be overidentied, with the number of instruments strictly larger than the number of endogenous variables. TABLE 12.1 Two Stage Least Squares Estimates of the Demand for Cigarettes Using Panel Data for 48 U.S. States Dependent variable: In(Q 1995 cigarettes) - In(Q1985 Regressor (1) (2) (3) In(pcigarettes) - In( pcigarettesy -0.94** -1.34** -1.20** 1,1985 (0.21) (0.23) (0.20) In(Inc;1995) - In(Inc;, 1985) 0.53 0.43 0.46 (0.34) (0.30) (0.31) Intercept -0.12 -0.02 -0.05 (0.07) (0.07) (0.06) Both sales tax and Instrumental variable(s) Sales tax Cigarette-specific tax cigarette-specific tax First-stage F-statistic 33.70 107.20 88.60 Overidentifying restrictions 4.93 J-test and p-value (0.026) These regressions were estimated using data for 48 U.S. states (48 observations on the 10-year differences). The data are described in Appendix 12.1. The J-test of overidentifying restrictions is described in Key Concept 12.6 (its p-value is given in parentheses), and the first-stage F-statistic is described in Key Concept 12.5. Individual coefficients are statistically significant at the *5% significance level or *$1% significance level