Table 4. Results of GLMs in relationship of the confirmed cases over population size per 100000 at every 5 consecutive days (CCOPSPTT) with socio-economic determinants (GDP and population), spatial distance (DW and DLPC), climate factors (relative air humidity, solar radiation intensity and average temperature) and time in cities outside and within Hubei province.
Types | Regression parameters | Baseline model | Time-delay model | ||||
---|---|---|---|---|---|---|---|
Coefficients | Coefficients | ||||||
Estimate | SE | p-value | Estimate | SE | p-value | ||
Hubei province | Intercept | -0.02 | 0.01 | 0.28 | -0.02 | 0.01 | 0.31 |
Population (Ten thousand) | -1×E-5 | 9×E-6 | 0.20 | -1×E-5 | 9×E-6 | 0.23 | |
GDP (Hundreds of millions RMB) | 5×E-6 | 6×E-7 | <0.001*** | 4×E-6 | 5×E-7 | <0.001*** | |
DW (km) | -1×E-5 | 1×E-5 | 0.25 | -1×E-5 | 1×E-5 | 0.23 | |
Relative air humidity (%) | 2×E-4 | 2×E-4 | 0.11 | 1×E-4 | 2×E-4 | 0.12 | |
Solar radiation intensity (h) | 5×E-4 | 4×E-4 | 0.27 | 4×E-4 | 4×E-4 | 0.26 | |
Average temperature (°C) | -1×E-3 | 6×E-4 | 0.06 | -1×E-3 | 6×E-4 | 0.07 | |
Time (d) | 4×E-4 | 3×E-4 | 0.11 | 3×E-4 | 3×E-4 | 0.12 | |
AIC | -635.68 | -643.23 | |||||
Marginal R2 | 0.52 | 0.51 | |||||
Conditional R2 | 0.52 | 0.51 | |||||
Other provinces | Intercept | 5×E-5 | 9×E-6 | <0.001*** | 4×E-5 | 9×E-6 | <0.001*** |
Population (Ten thousand) | 9×E-8 | 1×E-8 | <0.001*** | 8×E-8 | 1×E-8 | <0.001*** | |
GDP (Hundreds of millions RMB) | 1×E-8 | 9×E-10 | <0.001*** | 1×E-8 | 1×E-9 | <0.001*** | |
DW (km) | -2×E-8 | 6×E-9 | <0.01** | -2×E-8 | 6×E-9 | <0.01** | |
DLPC (km) | -6×E-9 | 2×E-8 | 0.75 | -5×E-9 | 2×E-8 | 0.75 | |
Relative air humidity (%) | -4×E-8 | 3×E-8 | <0.05* | -4×E-8 | 3×E-8 | <0.05* | |
Solar radiation intensity (h) | 3×E-6 | 8×E-7 | <0.001*** | 4×E-6 | 7×E-7 | <0.001*** | |
Average temperature (°C) | 4×E-7 | 3×E-7 | <0.05* | 3×E-7 | 3×E-7 | <0.05* | |
Time (d) | -4×E-6 | 3×E-7 | <0.001*** | -5×E-6 | 4×E-7 | <0.001*** | |
AIC | -32617.20 | -32634.55 | |||||
Marginal R2 | 0.33 | 0.32 | |||||
Conditional R2 | 0.33 | 0.32 |
The introduction of DW and DLPC see Table 1. The robustness of the model was tested by the similarity of regression coefficients between the baseline and time-delay models. If the regression coefficients are close to each other between the two models, indicating that our model has high robustness.