This PR adds in the epinowcast nowcasting model as a joint approach for estimating primary incidence and delay distributions (as a forwards process). I will add in the updated figures to the draft as well as adding a model outline to the methods (as a first draft).
Something that surprised me here was the bias towards over estimating the mean at short delays. I am not at all clear where that is coming from so thoughts appreciated.
This PR adds in the
epinowcastnowcasting model as a joint approach for estimating primary incidence and delay distributions (as a forwards process). I will add in the updated figures to the draft as well as adding a model outline to the methods (as a first draft).Something that surprised me here was the bias towards over estimating the mean at short delays. I am not at all clear where that is coming from so thoughts appreciated.