17a sir numberphile refined

I saw a numberphile video last night that simulated the sir model and it’s outcome looks a bit different to mine. The same with the one on wikipedia, it has a bit of a different trajectory when compared to mine. Here I explain that difference.

It really is due to mine being discreet (vs numberphile’s continuous one) in time and having a few important refinements (such as non-zero fatalities and non-instant recoveries) as discussed in detail in this video.

As a result my outcome sees a rapid decline in new infections whereas the continuous model shows a slow one - but see the video for more on this.

Wikipedia: https://en.wikipedia.org/wiki/Compartmental_models_in_epidemiology

Numberphile: https://youtu.be/k6nLfCbAzgo

Note this is the second version of fdm 17. The video is the same but around 50s of audio, from around 4:35, is replaced.