search

CDCgov / multisignal-epi-inference

Python package for statistical inference and forecast of epi models using multiple signals
https://cdcgov.github.io/multisignal-epi-inference/
10 stars 2 forks source link

Generation interval pmf #11

Open ghost opened 4 months ago

ghost commented 4 months ago

Goal

We want to have a single source of truth for the mathematical formulation of the generation interval probability mass function for flu and COVID. We want to be able to use this to immediately code up the math decided upon here.

Context

This was discussed in the 2024-02-27 meeting (see minutes) and agreed upon in the ADR record under model_features.md. This specific feature should be the same as what's implemented in the existing wastewater model. Completion of this task will allow us to implement the feature more quickly and should facilitate creation of a model design document.

Required features

Out of scope

Related documents

Ref: https://github.com/cdcent/cfa-multisignal-renewal/issues/72 author: @kaitejohnson

ghost commented 3 months ago

Hey @SamuelBrand1, here is one you could take! I also suggest: #17!

gvegayon commented 3 months ago

Hey @SamuelBrand1, I know the issue says "[t]his does not include code, just specification of the feature in math/writing," but I would argue that it is a good idea to include coding. In my view, coding stuff helps with the thinking process of theory too, and it makes sense to include the coding and documentation at the same time. What do you think?

cc @dylanhmorris

gvegayon commented 1 month ago

As a reference to https://github.com/cdcent/cfa-multisignal-renewal/discussions/104#discussioncomment-9374461, this issue deals with the implementation of a stochastic generation interval (i.e. for inference). The GI used in the hospital admissions-only version of the WW project uses a deterministic quantity.

cc @kgostic

kgostic commented 1 month ago

Just to clarify about the generation interval: you're saying that for now, you're not going to try to estimate the GI in the model. Instead, you're just going to pass in a vector $\vec{v}$ where entries represent the probability mass of the discretized GI for integer delays of $d$ days.

If so, this is also how we're doing it in our $R_t$ model, so I think this is totally sufficient for now.