Fig. 2

The extended working day moving average applied to synthetic data, with the seven-day and working day moving averages for comparison. Synthetic data were generated for 28 days, containing day of the week and public holiday effects representative of those observed in the GP in-hours syndromic surveillance system, but without noise and longer term trends. The synthetic data included a public holiday Monday. This is indicated by the grey vertical line and easily identifiable by the negligible number of consultations on this day. The extended working day moving average was applied to this data with the seven-day and working day moving average shown for comparison. The red box highlights the pre- and post- public holiday period of interest