Mahesh Ramanan

60 Followers
61 Following
117 Posts
Intensivist with an interest in clinical trials and ICU databases
Amateur gardener
Cricket fanatic

ICYMI this paper really is a must read: https://doi.org/10.1111/ele.14033

Comparing and choosing best models based on AIC etc, then interpret coefficients causally (what's the effect of X on Y?) is flawed, yet so common

We must draw causal assumptions first (DAG)

"Model selection is not a valid method for inferring causal relationships. It's appropriate for predictive inference (which model best predicts Y?), which is fundamentally distinct from causal inference (what is the effect of X on Y?)"

1/

Andrew Gelman on variable selection: "Variable selection (that is, setting some coefficients to be exactly zero) can be useful for various reasons, including:
– It’s a simple form of regularization.
– It can reduce costs in future data collection.
Variable selection can be fine as a means to an end. Problems can arise if it’s taken too seriously, for example as an attempt to discover a purported parsimonious true model."

https://statmodeling.stat.columbia.edu/2023/07/18/when-your-regression-model-has-interactions-do-you-need-to-include-all-the-corresponding-main-effects/#comments

Hear, hear. Saving this for future use.

When your regression model has interactions, do you need to include all the corresponding main effects? | Statistical Modeling, Causal Inference, and Social Science

@LucillaPiccari @gpollara

Your “naivety as young researcher” 😂 made me think…
What if labs were “federated” in ressources ? Statisticians, ressources, equipment, data, etc all accessible to a “hub” of ressources for researchers.
This would certainly help cut cost between researchers or optimize ressources for public funded researchers. It could stimulate research & science 🤔

I’m kinda brainstorming here 😅

Musk revokes Twitter accounts for transit agencies, as they would compete against Tesla. Americans call this Free Markets.

https://octodon.social/@ethanz/109683937494252761
[email protected] - As Twitter suspends city services running on its infrastructure, maybe the question we should ask is "Why did we trust our services to privately owned platforms in the first place?" Excellent piece by @[email protected] of Bloomberg: https://www.bloomberg.com/news/articles/2023-01-13/despite-chaos-cities-are-sticking-with-twitter?sref=ndQeXqOs

Ethan Zuckerman (@[email protected])

As Twitter suspends city services running on its infrastructure, maybe the question we should ask is "Why did we trust our services to privately owned platforms in the first place?" Excellent piece by @[email protected] of Bloomberg: https://www.bloomberg.com/news/articles/2023-01-13/despite-chaos-cities-are-sticking-with-twitter?sref=ndQeXqOs

the Octodon

Still getting the hang of Mastodon. Hope to use it more in the new Year for conferences and #foamed.

In the meantime- Happy New Year, friends. 🥂

Today's Osler Top 5 from 2022 - Articles

Here's the year's best articles as voted by our members

#medmastodon #meded #medtwitter #FOAMed #FOAMcc #emergencymedicine #juniordoctors #tipsfornewdocs #intensivecare #ICU #emergencyphysician

@mindfullmedic

Who are the #epidemiologists and infectious disease #modellers on Mastodon?
Will update this list as I come across them & if you “favourite” or “boost” this toot you will be updated every time there is an addition.
This helps create an “Epidemiology/Modellers” list on mastodon to browse their focused stream.
List so far:
@famulare
@enenbee
@aetiology
@DrZoeHyde
@dgbassani
@doxy_cycling
@adamjkucharski
@dgurdasani1
@epiellie
@mlipsitch
@deonandan
@seabbs
#epidemiology #epidemiologist

An overview of the ACCP training pathway in the UK:

https://www.sciencedirect.com/science/article/abs/pii/S1036731422002508?dgcid=author

We really need our colleges and professional bodies to look into this in Australia.

This is why they need to claim universities are corrupt.

@EposVox As an addendum to this: most authors are not only genuinely delighted to share their work with you if you request it, some are even moreso delighted to talk to you about it and answer questions!

Especially if they have additional knowledge or information worth sharing that they didn't include in the original paper!