Table 8.

Bias of estimates of $R_0$ for DGP (5.5)–(5.6).

	Deaths	Cases	SUR, Efficient $W$	SUR, Diagonal $W$	SUR, Identity $W$
Panel A: No Rounding, True $E_0=2$ and $I_0=0$, Assumed $E_0=2$ and $I_0=0$
Levels, Cumulative	0.0007	0.0004	0.0009	0.0005	0.0002
Levels, Daily	0.0078	−0.0027	0.0026	0.0031	−0.0013
Logs, Cumulative, Trim	0.0003	−0.0013	−0.0007	−0.0009	−0.0005
Logs, Daily, Trim	0.0011	−0.0001	0.0012	0.0011	0.0004
Logs, Cumulative, No Trim	0.0002	0.0003	0.0002	0.0002	0.0002
Logs, Daily, No Trim	0.0001	0.0002	0.0002	0.0002	0.0002
Panel B: No Rounding, True $E_0=2$ and $I_0=0$, Estimated $E_0$ and $I_0$
Levels, Cumulative	0.0020	0.0014	0.0040	0.0054	0.0110
Logs, Cumulative, Trim	0.0010	−0.0006	0.0025	0.0040	0.0001
Logs, Cumulative, No Trim	0.0029	0.0047	0.0022	0.0022	0.0025
Panel C: Rounding, True $E_0=2$ and $I_0=0$, Assumed $E_0=2$ and $I_0=0$
Levels, Cumulative	0.081	0.005	0.016	0.021	0.005
Levels, Daily	0.017	−0.003	0.007	0.008	−0.001
Logs, Cumulative, Trim	0.124	0.006	0.029	0.034	0.065
Logs, Daily, Trim	0.023	0.001	0.010	0.010	0.012
Logs, Cumulative, No Trim	0.669	0.038	0.017	0.040	0.349
Logs, Daily, No Trim	−0.271	0.005	−0.005	−0.002	−0.134
Panel D: Rounding, True $E_0=2$ and $I_0=0$, Estimated $E_0$ and $I_0$
Levels, Cumulative	0.082	0.006	0.022	0.033	0.015
Logs, Cumulative, Trim	0.125	0.007	0.030	0.038	0.066
Logs, Cumulative, No Trim	0.671	0.038	0.018	0.041	0.350

The table presents the average bias of the estimates of $R_0$ across 500 simulation draws for different estimation methods. The estimates in the first column are based on the deaths data only. The estimates in the second column are based on the reported cases data only. The estimates in the remaining three columns combine the deaths and reported cases data and use different weighting matrices.