Table 4.

The Log-linear panel regression of weekly new confirmed cases on state-level factors [Depended variable: log(D(cases))]

Explanatory variables	Model 1	Model 2	Model 3	Model 4
Constant	− 0.492(− 0.911)	− 0.467(− 1.691)	0.816*(2.110)	− 3.560**(− 4.412)
Log(D(Cases(− 1))	0.785**(34.931)	0.796**(23.556)	0.774**(21.010)	0.779**(21.745)
Log(PCNSDP)				0.234**(3.925)
Log(URBAN)	0.149(1.540)		0.156*(2.059)
Log(DENSITY)		0.070**(2.932)		0.073**(3.005)
Log(AGRI)	− 0.075*(− 2.011)		− 0.079(− 1.834)
Log(BPL)	− 0.079(− 1.801)	− 0.087(− 1.770)	− 0.083(− 1.724)
TIME	0.132**(4.919)	0.122**(4.820)		0.138**(5.099)
Time-squared	− 0.002**(− 5.497)	− 0.002**(5.430)		− 0.003**(− 5.668)
R-squared	0.91	0.91	0.91	0.91
Adjusted R-squared	0.91	0.91	0.91	0.91
F-statistic	2539.62**	3027.57**	3041.57**	3062.94**

Source: Estimated by the authors on the basis of secondary data

(1) Numbers in the parentheses are t ratios which use White’s diagonally corrected standard errors throughout. **means significant at 1% level and *means significant at 5% level. (2) Number of states and UTs = 33, number of weeks = 47; panel includes 1551 pooled observations. (3) D(cases) imply the first difference of cumulative weekly total cases, which is tantamount to weekly count per million or weekly new cases per million. (4) Model 3 is estimated under fixed time effects (periods fixed) supressing time and time-squared