模型存在内生性,需要用工具变量处理。
做如下检验:
1、模型确实存在内生性检验。DWH。原假设:模型中所有变量均为外生的,或者说不需要使用工具变量了。
2、工具变量和内生变量有可观的相关性——识别不足检验?。F检验、偏R2检验。原假设:第一阶段回归中,所用的工具变量对内生变量无解释力。
3、工具变量本身是外生的——过度识别检验(只能在工具变量大于内生变量数目时使用)。sargan检验(原假设:所有工具变量均为外生的),hansen检验(原假设是什么?)另外,下面的动态回归结果,可以接受吗?sargan和hansen应该看哪个?两个hansen应看哪个?多谢!
| Arellano-Bond test for AR(1) in first differences: z = -5.17 Pr > z | = | 0.000 |
| Arellano-Bond test for AR(2) in first differences: z = -0.26 Pr > z | = | 0.797 |
| |
| Sargan test of overid. restrictions: chi2(80) = 144.39 Prob > chi2 | = | 0.000 |
| (Not robust, but not weakened by many instruments.) |
| Hansen test of overid. restrictions: chi2(80) = 93.06 Prob > chi2 | = | 0.151 |
| (Robust, but can be weakened by many instruments.) |
| Difference-in-Hansen tests of exogeneity of instrument subsets: |
| iv(yr1976 yr1977 yr1978 yr1979 yr1980 yr1981 yr1982 yr1983 yr1984) |
| Hansen test excluding group: chi2(73) = 84.71 Prob > chi2 | = | 0.164 |
| Difference (null H = exogenous): chi2(7) = 8.35 Prob > chi2 | = | 0.303 |