Methods
Linear regression models were calculated to assess and quantify the
effects of general testing activity and public mobility on the number of
confirmed cases per day. Pearson correlation coefficients were
calculated to further explore the type of correlation between mobility
and number of confirmed cases.
To visualize a potential impact of major political decisions on the
trend in newly confirmed cases, residuals were obtained from the linear
regression model including background test activity. The residuals were
submitted to time series decomposition to examine seasonal variation per
calendar week and the remaining error variation. The trend curve,
defined as moving average over seven days, was used to identify trend
changes in plausible temporal relationship to tightening or loosening of
governmental restrictions.