Baking Even Better With ARIMA (part 2)

Assessing the influence of mass viral testing programmes on case positivity: are we experiencing a testdemic? (rev 1.0)

In part 1 of this series I conjured an ARIMA model for the prediction of new first episodes by specimen date and out popped unique people tested and LFD tests as highly statistically significant predictors of new first episodes (p<0.001 in each case). This places the emphasis on the blasted LFD test as a testdemic generator, which isn’t going to surprise anybody!

This morning I fancied one of those all-in-the-pot approaches whereby we flip back to using new daily cases by specimen date as the dependent series (i.e. first episodes + reinfections) and throw in daily counts of both LFD and PCR tests as well as the peculiar unique people tested series.

I turned the handle on this jam-packed sandwich and a few seconds later this was thrown back at me…