Benjamin Shaya

Ah, yes, prior probability vs posterior probability. Let's say that, today, I decide that I want a 1% chance of getting COVID over the next year. I decide to spend...

I don't think the average person has intuition on how many million sided die rolls they should tolerate for a given activity. On Thu, Jan 27, 2022 at 4:56 AM...

It looks like update_prevalence.py needs to roll down vaccine numbers from nations to states outside the US (within the US we get state-level vaccination rates). I'm not sure if it's...

To fix: Update rolldown_vaccine_types to copy the # of vaccinations, scaled by population from parent location to child location: https://github.com/microcovid/microcovid/blob/main/update_prevalence.py#L645 Call rolldown_vaccine_types on countries: https://github.com/microcovid/microcovid/blob/main/update_prevalence.py#L654 Run update_prevalence

I've removed the selector when there isn't data. I'll try to get around to using Country-level data when region data isn't available (better to do this in the FE than...

@Kathy, Sorry for the flipflopping, I think this should be done one the frontend. On Mon, Sep 20, 2021 at 9:59 AM Jenny ***@***.***> wrote: > I think this is...

Thanks for the studies, definitely stuff for me to chew on. The paper we cited was https://www.imperial.ac.uk/media/imperial-college/medicine/mrc-gida/2021-12-16-COVID19-Report-49.pdf. The Denmark study blends all "boosted" vaccines, which includes Johnson&Johnson, which has a...

Good point on these registries not capturing differences in test-seeking. The paper you cited suggests that case-negative studies would overcompensate for this, though: > An expected consequence of this design...

It's pretty hard to review this with all the csv files in the same commit as the code changes. Any chance you could remove the csv files from this pull...

Hm, good point. I suppose that if a filter is 85% efficient at capturing the particles, you could treat it as 85% of a hepa filter. I.e. to get 5...