Pausal Živference @PausalZ@qoto.org
- GitHub
- https://github.com/pzivich
- website
- https://pzivich.github.io/
Paul Zivich. Computational epidemiologist, causal inference researcher, and open-source enthusiast #epidemiology #statistics #python
Joined Apr 2022
@healthstatsdude another way would be to fit/report regression models separately by race/ethnicity
More causal then regression, but I really enjoy the following paper by John Jackson https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8478117/
Pausal Živference
boosted
Two wishes:
1. For people to stop trashing and misrepresenting about RCTs to justify doing anything else RCTs
2. For to stop with the lazy, facile 'RCTs are the gold standard', which is why wish (1) is needed
the format for JAMA's abstracts are such a pain
Pausal Živference
boosted
Hi, my name is Maarten. You might know me from Twitter from my greatest hits "you don't nee-hee-hee-heed machine learning for that" and "gimme, gimme, gimme your research question"
@PWGTennant it just has a nice ring to it. also Z-estimators are a real thing, so might be doubly difficult
@PWGTennant RE: slow, that depends on the instance and their server availability. I think social is probably a bit slower since most people seem to have migrated there first.
Pausal Živference
boosted
I'm excited that #ScienceTwitter & #AcademicTwitter are flocking to #Mastodon as the 'replacement for Twitter'.
It may be confusing, clumsy, & slow, but it's free, open-source, & absolutely not-for-profit. That's philosophically far more appropriate. I hope we can make it work!
@statsepi as long as it gets clicks, "journalists" get to write whatever they want
an easy way to spice up your statistical analysis #epidemiology #biostatistics
@tfeend
RE: finding people. I think it's a higher start-up cost. There isn't an algorithm to fill out your feed, so it requires more effort on the user. I found looking through the followers of people I know an easy way to start filling out my home feed
@tfeend the speed depends on which instance you are on. my guess is that mastodon.online has had a large influx past their current server capacity.
@rwidome but to look on the bright side, I think there is an opportunity to build a strong community and have some of the epitwitter glory days (like before things got a little off the rails in 2020 lol)
@rwidome my biggest gripe is how the reply threads are displayed. I wish they had a nested display
@kaz_yos my instance gives me 65k characters lol. I could put a whole paper if I wanted to
@kaz_yos I think #epidemiology makes the most sense
@rwidome I mean I think #epidemiology makes the most sense (and easiet to find), but I know people like to mix epi with something platform specific. While they haven't caught on, I've seen #epitoots #epitodon
@rwidome yeah, your feed has to be 'tailored' by you a bit more. I've found looking through followers of others to be a bit helpful. Really, a central hashtag needs to be adopted by the epi's
The other thing is boosting is more important than favoring (since you don't get 'served' popular tweets)
As with everything, there is always the start-up costs
@rwidome I don't think you have to start from scratch (see link below). I may wait a little to see where most of the epi's go (since there's a long cooldown between switching instances)
Lately, M-estimation has been my go-to tool for analyses. It's just such a flexible tool and makes variance estimation super easy
I did a talk a few weeks back, the recording is available at
https://lshtm.cloud.panopto.eu/Panopto/Pages/Viewer.aspx?id=9068252c-3aa1-4a26-a9a0-af2d00d802b1
with slides and code here
- GitHub
- https://github.com/pzivich
- website
- https://pzivich.github.io/
Paul Zivich. Computational epidemiologist, causal inference researcher, and open-source enthusiast #epidemiology #statistics #python
Joined Apr 2022
