Table 2.
Estimation results of the effects of PM2.5 on the total medical costs
| Variables | ln (total medical costs) | Total medical costs | |||
|---|---|---|---|---|---|
| (1) | (2) | (3) | (4) | (5) | |
| Linear regression | Fixed-effects 2SLS | Pool-Tobit | Tobit-CRE | Tobit-CRE-CF | |
| PM2.5 | − 0.008*** | 0.052 | − 16.527*** | − 47.180* | 526.396* |
| (0.001) | (0.048) | (4.598) | (18.675) | (232.711) | |
| PM2.5_residual | - | - | - | - | − 576.490* |
| - | - | - | - | (230.521) | |
| Other control variables | Yes | Yes | Yes | Yes | Yes |
| Year dummy | Yes | Yes | Yes | Yes | Yes |
| Correlated random effects | No | No | No | Yes | Yes |
| N | 26,784 | 26,784 | 26,784 | 26,784 | 26,784 |
| R2 | 0.342 | 0.180 | - | - | - |
| Cragg-Donald Wald F statistic | - | 48.229 > 19.93 | - | - | - |
| Sargan statistic | - | 1.245 (P = 0.265 > 0.05) | - | - | - |
Note: (1) Standard errors in parentheses; (2) Other control variables include a set of covariates; (3) Correlated random effects: a linear combination of endogenous variables, instrumental variables and covariates; (4) *p < 0.05, **p < 0.01, ***p < 0.001