Consider the Baltagi and Levin (1986) cigarette demand example for 46 states. This data, updated from 1963-92, is available on the Springer web site as cigar.txt.

(a) Estimate equation (8.43) using 2SLS, FD-2SLS and their Keane and Runkle (1992) version. (Assume only \(\ln C_{i, t-1}\) is endogenous).

(b) Estimate question (8.43) using the Within and FE-2SLS and perform the Hausman-type test based on FE-2SLS versus FD-2SLS.

(c) Perform the Hausman-type test based on 2SLS versus FD-2SLS.

(d) Perform the Anderson and Hsiao (1982) estimator for equation (8.43).

(e) Replicate the two-step robust Arellano and Bond (1991) GMM estimator for equation (8.43) given in Table 8.2.

(f) Replicate the System GMM estimator for equation (8.43) given in Table 8.3.

Table 8.2:

Table 8.3:

\[\begin{equation*}
\ln C_{i t}=\alpha+\beta_{1} \ln C_{i, t-1}+\beta_{2} \ln P_{i, t}+\beta_{3} \ln Y_{i t}+\beta_{4} \ln P n_{i t}+u_{i t} \tag{8.43}
\end{equation*}\]

xtabond2 Inc L. (Inc) inrp inrpn Inrdi dum3 dumb dum10-dum29, gmm (L. (Inc), collapse) iv (Inrp Inrpn Inrdi dum3 dum8 dum10-dum29) noleveleq robust nomata twostep Dynamic panel-data estimation, two-step difference GMM Group variable: state. Time variable yr Number of instruments - 53 Wald chi2 (26) Prob > chi2 5492.89 0.000 Number of obs 1288 Number of groups 46 Obs per group: min 28 avg 28.00 max 28 Corrected Coef. Std. Err. [95% Conf. Interval] Inc | Ll. .6993342.0688738 10.15 0.000 .5643441 .8343242 Inrp .3956542 .0659587 -6.00 0.000 -.5249309 -.2663775 1nrpn 1054687 .0846056 -1.25 0.213 -.2712926 .0603552 Inrdi | .1258248 .0359738 3.50 0.000 .0553173 .1963322 Arellano-Bond test for AR (1) in first differences: z = Arellano-Bond test for AR (2) in first differences: 2 - -4.74 1.82 Pr > z = 0.000 Pr2 0.069 32.25 Prob > chi2 0.223 Hansen test of overid. restrictions: chi2 (27) Warning: Sargan/Hansen tests are weak when instruments are many. Time dummies are not shown here to save space..