search

epiforecasts / inc2prev

Estimate epidemiological quantities from repeated cross-sectional prevalence measurements
http://epiforecasts.io/inc2prev/
BSD 3-Clause "New" or "Revised" License
16 stars 7 forks source link

Antibody model fix #27

Closed sbfnk closed 2 years ago

sbfnk commented 2 years ago

It seems the factors are the wrong way around in the antibody model. A reasonable assumption is:

1) Infections occur in those without antibodies, such raise the absolute level of antibodies without adjustment 2) Vaccinations are given randomly in the population, and thus the probability of ending up in people without antibodies is equal to the population proportion without antibodies.

This PR reflects this assumption.

sbfnk commented 2 years ago

Alternative is to let the model decide (implemented here).

seabbs commented 2 years ago

👍🏻