Peak of the number of daily deaths caused by Covid19

[kent davidge]

Around the mid of april, I had done some calculations to get some numbers for daily deaths in my country due to covid19. It was really simple, just a gaussian without taking into account any other factor.

I concluded that if the peak of daily deaths occurred on april 30, we would have about 500 deaths that day. The officially reported number of deaths turned out to be about 500! So if my model was correct, we already passed the peak, right?

But how to know if my model was correct? It seems that I have to wait until the "end" of the pandemic, to see if the peak was indeed reach on april 30. Correct? If that's the case, then my model was giving me the right numbers.

 

[FactChecker]

Or you were lucky. It would be nice if there was some theoretical motivation for your model. This is especially true when you are dealing with something like a virus epidemic, where the growth can be exponential and there has been a lot of work on models that have a lot of logic behind them.

 

[kent davidge]

Or you were lucky

It would be nice if there was some theoretical motivation for your model

I think the main motivation is that there must be an increase as more and more people get infected, then as less people get infected (since a large portion of the population already got the virus) less people will die? Until we reach a point far away, aka letting time go to infinity, where essentially nobody is dying from the virus.

I think the most simple mathematical function for describing that is a gaussian.

 

[jim mcnamara]

The SIR model has a compartmental approach - susceptible, ill, recovered(dead included)
- which sounds a little like what you may have done.
This article discusses compartmental models, see what you can do with them. You already have a data set.

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