. 2021 Dec 18;15:e01076. doi: 10.1016/j.sciaf.2021.e01076

Table 5.

Regression results.

Variable	Pre-Epidemic	Epidemic	Pandemic
Variable	Pre-Epidemic	Epidemic	Cases	Deaths
$e r_{i t}$	0.0174** (0.0077)	0.1550 (0.1630)	−6.22e-05 (0.0007)	7.86e-05 (0.0013)
$o p_{i t}$	0.0054 (0.0049	−0.0024 (0.0132)	−3.66e-05 (0.0001)	−7.67e-05 (0.0006)
$t c c_{i t}$			−1.34e-05 (0.00412)
$t c d_{i t}$				−6.17e-05 (8.37e-05)
$Δ e r_{i t}$	0.0226*** (0.0051)	−0.0303** (0.0153)	−0.0184 (0.0431)	−0.0171 (0.0606)
$Δ o p_{i t}$	−0.0114 (0.0165)	−0.0092** (0.0036)	−0.0073*** (0.0028)	−0.0104 (0.0069)
$Δ t c c_{i t}$			−0.0076** (0.0029)
$Δ t c d_{i t}$				0.0010 (0.0039)
$E C T_{i t - 1}$	−0.8501*** (0.0658)	−0.8090*** (0.0930)	−0.9040*** (0.0563)	−0.9330*** (0.0531))
$C o n s t a n t$	0.0012 (0.0010)	0.1040*** (0.0264)	−0.0170*** (0.0039)	−0.0095** (0.0048)
Hausman test $(χ_{k}^{2})$	2.10(0.3496)	5.22(0.0736)	6.23(0.1011)	1.84(0.6054)
Log likelihood	35,843.54	3225.81	9676.76	8527.46
No. of cross-sections	12	12	12	12
Number of observation	15,372	780	2691	2446

Note: The values in parenthesis are the standard errors for the estimates but probability values for the Hausman test while ***, ** & * implies significance at 1%, 5% and 10%, respectively. For the Hausman test, the PMG estimator is the efficient estimator under the null while the MG estimator is the efficient estimator under the alternative hypothesis.