Table 5.
Benchmark regression results.
| Variable | Dependent variable: CTFP | ||
|---|---|---|---|
| (1) | (2) | (3) | |
| Estimation method | OLS | RE | FE |
| Constant | −0.050 (−0.34) | 0.052 (0.23) | 7.456** (2.18) |
| LnDIG | 0.344*** (12.99) | 0.330*** (13.89) | 0.319*** (12.37) |
| LnPGDP | 0.056*** (5.40) | 0.045*** (4.31) | 0.041*** (3.69) |
| LnPOP | 0.061 (1.40) | 0.063 (0.78) | −3.030** (−2.56) |
| LnPOP2 | −0.004 (−0.99) | −0.004 (−0.51) | 0.299*** (2.83) |
| LnENVR | −0.035* (−1.72) | −0.059*** (−2.60) | −0.065*** (−2.65) |
| FDI | −1.936** (−2.25) | −3.074*** (−3.19) | −3.131*** (−2.96) |
| LnST | 0.081*** (2.93) | 0.033 (1.10) | 0.011 (0.34) |
| R2 | 0.584 | 0.675 | 0.687 |
| OBS | 300 | 300 | 300 |
| F statistic | F(29, 263) = 6.11, Prob>F = 0.0000 | ||
| Hausman tests | chi2(8) = 55.64, Prob>F = 0.0000 | ||
The values in parentheses are t-values; *, **, and *** indicate that the estimated coefficients are significant at the 10, 5, and 1% levels, respectively. OLS, RE, and FE represent mixed effects estimation, random effects estimation, and fixed effects estimation, respectively. The asterisks in the following tables have the same meaning as the notes here.