Supplementary Material for Dynamic COVID risk assessment accounting for community virus exposure from a spatial-temporal transmission model

Gaussian distribution that the inverse of the Gaussian covariance matrix is the partial correlation. These are all publicly available databases. In Figure D.1, we plot the covariates used in the disease In Figure D.2, we plot the estimated infection D.5, we show the additional results for the community-level COVID transmission model in estimating Here we describe the procedure to construct the confidence intervals for the parameters in the spatiotemporal model. We did not permute across areas as it might disturb spatial correlation. Figure D.3: Rooted mean squared errors (RMSEs) in estimating the time-varying parameters in the RMSE value was calculated over all areas and time points in each replication.