COVID-19 AU: Predictions using 7-Day Moving Averages

In this post we look at using 7-day moving averages to estimate the midpoint and total number of confirmed COVID-19 cases in Australia.

A 7-day moving average adds together the number of new cases for the last seven days then divides the total by seven.

Below is a bar-chart for Australia confirmed cases:

Below we have a graph of the 7-day moving averages and the 7-day moving averages for these:

Once again the moving averages of the moving averages (MAMA) is used to estimate the lag time.

The midpoint occurs in the dip in the blue curve on 30 March.

The red curve peaks around 1-2 April.

We look at the bar Chart at 28 March and see a low point (212) for the number of cases.
(Date, Cumulative confirmed cases, Daily cases)

The cumulative number of cases for 28 March is 3378 cases.

We double this (since the low point is for a single date) to get an estimated total number of cases of 6756.

We need to as usual add on 10% to allow for the tail, giving a total of 7431 cases (say 7500 cases).

These are the latest results:
(Date, Cumulative confirmed cases, Daily cases)

We look at the bar chart again and see we are already in the tail.

We anticipate the total number of cases is in the range of 7000-7500 cases.

We could have generated the above range a month ago using the above method.

The method appears to be working for Australia and New Zealand.

Green Bottles, Alan's Ark & COVID Odyssey: Alan Grace's vir[tu]al journey