An Enigma: Transmission Of Epidemic Influenza (part 10)

I attempt to shed light on the riddle that is seasonal influenza using my bag of spanners. Today I wield the spanner of ARIMA and bring it down on mortality estimates for females

Well here we all are washed, trimmed and ready for the second spannering. It will make terribly good sense to go back and read part 9 so you can see where I’m coming from with this because I’d like to plough straight in by revealing the nine ARIMA models for female mortality. And here they are, as revealed one table at a time…

Models For Female Influenza Mortality

We start once again with a summary table of model specification. Please note the model number for each age band because you’ll need to use this to identify graphical outputs later on. The model type is given as ARIMA(p,d,q), with ‘p’ representing the order of autoregressive terms (AR), ‘d’ representing the order of series differencing, and ‘q’ representing the order of the moving average terms (MA).

Note once again the similarity among the model structures, with ARIMA(0,1,1) grabbing the gold with 6 out of 9. Not also the similarity with the structure matrix for males, with 6 fetching-up as identical. Slight differences are to …