From: The forecast of COVID-19 spread risk at the county level
Accuracy of the COVID-19 forecasting model | Significant testing correlation 0f 0.83 (p = 0.0053) with MAE of 605.4 accumulated cases |
---|---|
Correlation analysis for individual counties | The majority of the counties had a correlation of>0.7 across the states |
Relationship between model performance and number of cases | The county number of cases did not affect the model performance. |
Capturing change in COVID-19 cases due to government responses | The model captures the decrease in the cases with higher stringency index |
Ability to capture the effect of age demographics on COVID-19 cases | Average daily cases has reverse proportionality with the retire percentage and direct proportionality with the young population percentage |