Zheer Kejlberg Al-Mashhadi
@zheer.bsky.social
MD, PhD in endocrinology and epidemiology.
Postdoc @ Steno Diabetes Center Aarhus
Interested in (pharmaco)epi, stats, causal inference. Trying to better characterise complex medication use patterns — and their effects.
Postdoc @ Steno Diabetes Center Aarhus
Interested in (pharmaco)epi, stats, causal inference. Trying to better characterise complex medication use patterns — and their effects.
Reposted by Zheer Kejlberg Al-Mashhadi
what software do people use to generate pretty-looking causal DAGs for their papers? (answers that take the form of an R package preferred, but “we always use Powerpoint/Canva/… because doing it in code is too hard” is a useful data point too)
cc @pwgtennant.bsky.social @dingdingpeng.the100.ci
cc @pwgtennant.bsky.social @dingdingpeng.the100.ci
April 13, 2025 at 5:48 AM
what software do people use to generate pretty-looking causal DAGs for their papers? (answers that take the form of an R package preferred, but “we always use Powerpoint/Canva/… because doing it in code is too hard” is a useful data point too)
cc @pwgtennant.bsky.social @dingdingpeng.the100.ci
cc @pwgtennant.bsky.social @dingdingpeng.the100.ci
To all my econ friends:
On the plus side, a lot of brand-new instrumental variables are being created these days
On the plus side, a lot of brand-new instrumental variables are being created these days
April 9, 2025 at 8:43 PM
To all my econ friends:
On the plus side, a lot of brand-new instrumental variables are being created these days
On the plus side, a lot of brand-new instrumental variables are being created these days
the only reason I’m not smarter is that unrestricted comprehension would lead to paradoxes
March 29, 2025 at 2:56 PM
the only reason I’m not smarter is that unrestricted comprehension would lead to paradoxes
Pretendimand,
noun
The target of an analysis where you know your assumptions to be wrong yet assume them anyway.
noun
The target of an analysis where you know your assumptions to be wrong yet assume them anyway.
March 19, 2025 at 11:32 AM
Pretendimand,
noun
The target of an analysis where you know your assumptions to be wrong yet assume them anyway.
noun
The target of an analysis where you know your assumptions to be wrong yet assume them anyway.
I would read all those pdfs I have bookmarked, but my harddrive keeps overheating
March 15, 2025 at 7:20 AM
I would read all those pdfs I have bookmarked, but my harddrive keeps overheating
Reposted by Zheer Kejlberg Al-Mashhadi
The long list of words flagged for review includes several fundamental epidemiology terms which likely show up in every paper, grant, report, presentation, or conference talk.
Does this mean epidemiology as a whole is now forbidden?
My thoughts at: open.substack.com/pub/epiellie...
Does this mean epidemiology as a whole is now forbidden?
My thoughts at: open.substack.com/pub/epiellie...
March 12, 2025 at 2:45 PM
The long list of words flagged for review includes several fundamental epidemiology terms which likely show up in every paper, grant, report, presentation, or conference talk.
Does this mean epidemiology as a whole is now forbidden?
My thoughts at: open.substack.com/pub/epiellie...
Does this mean epidemiology as a whole is now forbidden?
My thoughts at: open.substack.com/pub/epiellie...
Reposted by Zheer Kejlberg Al-Mashhadi
Dive into stat modeling&applications, relaxing assumptions, predictive accuracy, preserving information, imputation, model validation, Bayesian models, causal specification+many controversial topics: Regression Modeling Strategies course May 15-16, 19-20 hbiostat.org/doc/rms/4day.html #StatsSky
RMS
hbiostat.org
March 9, 2025 at 5:41 PM
Dive into stat modeling&applications, relaxing assumptions, predictive accuracy, preserving information, imputation, model validation, Bayesian models, causal specification+many controversial topics: Regression Modeling Strategies course May 15-16, 19-20 hbiostat.org/doc/rms/4day.html #StatsSky
@noahgreifer.bsky.social still out there doing the Lord’s work
stats.stackexchange.com/questions/66...
stats.stackexchange.com/questions/66...
Should I perform double adjustment after propensity score weighting if all covariates are perfectly balanced?
I am using the WeightIt package and Covariate Balancing Propensity Score (CBPS) with link = logit to perform propensity score weighting and have achieved perfect balance (i.e., all absolute standar...
stats.stackexchange.com
March 5, 2025 at 12:09 PM
@noahgreifer.bsky.social still out there doing the Lord’s work
stats.stackexchange.com/questions/66...
stats.stackexchange.com/questions/66...
As a clinician-scientist working overtime to keep up with two lines of work, I’m so glad neither statistics nor medicine are plagued by poor naming practices and rampant use of eponyms.
January 25, 2025 at 8:53 AM
As a clinician-scientist working overtime to keep up with two lines of work, I’m so glad neither statistics nor medicine are plagued by poor naming practices and rampant use of eponyms.
Reposted by Zheer Kejlberg Al-Mashhadi
Well, it seems Aaron Siri sure didn't waste any time taking advantage of his NY Times notoriety to really push home the message that he is all for children's suffering.
Centuries ago, someone like him would be a Grimm's Fairy Tale character. 🛟😷 medsky health policy episky
Centuries ago, someone like him would be a Grimm's Fairy Tale character. 🛟😷 medsky health policy episky
On other social media sites this weekend there's a movement to "Make #Rubella Great Again." And we have data about morbidity and mortality due to rubella in the 1960s - #MedSky and #HistorySky.
December 14, 2024 at 9:29 PM
Well, it seems Aaron Siri sure didn't waste any time taking advantage of his NY Times notoriety to really push home the message that he is all for children's suffering.
Centuries ago, someone like him would be a Grimm's Fairy Tale character. 🛟😷 medsky health policy episky
Centuries ago, someone like him would be a Grimm's Fairy Tale character. 🛟😷 medsky health policy episky
The best way to deal with imbalanced datasets is to downsample the majority class.
Follow me for more machine learning tips and tricks
Follow me for more machine learning tips and tricks
December 13, 2024 at 9:40 AM
The best way to deal with imbalanced datasets is to downsample the majority class.
Follow me for more machine learning tips and tricks
Follow me for more machine learning tips and tricks
How do y’all cope with all this rejection in #academia? I’ve finally gotten way past that point in life where I take it personally, but it’s still just horrible and stressful and mildly angering
December 11, 2024 at 7:23 PM
How do y’all cope with all this rejection in #academia? I’ve finally gotten way past that point in life where I take it personally, but it’s still just horrible and stressful and mildly angering
Reposted by Zheer Kejlberg Al-Mashhadi
#rstats #quarto
Do you want to use different versions of a figure for HTML, PDF for online vs. print, perhaps an animated GIF for the former, but a static image for the latter?
Here's a discussion of this and solutions,
github.com/quarto-dev/q...
Do you want to use different versions of a figure for HTML, PDF for online vs. print, perhaps an animated GIF for the former, but a static image for the latter?
Here's a discussion of this and solutions,
github.com/quarto-dev/q...
How to provide alternative figures for HTML (.gif) and PDF (.png) versions of a book? · quarto-dev quarto-cli · Discussion #8172
Description Using Quarto: 1.3.450 In a book project I have several animated .gif images for the HTML version, but I need to replace these with static .png images for the LaTeX / PDF version. Follow...
github.com
December 5, 2024 at 8:59 PM
#rstats #quarto
Do you want to use different versions of a figure for HTML, PDF for online vs. print, perhaps an animated GIF for the former, but a static image for the latter?
Here's a discussion of this and solutions,
github.com/quarto-dev/q...
Do you want to use different versions of a figure for HTML, PDF for online vs. print, perhaps an animated GIF for the former, but a static image for the latter?
Here's a discussion of this and solutions,
github.com/quarto-dev/q...
This goes so hard
December 5, 2024 at 6:14 PM
This goes so hard
Reposted by Zheer Kejlberg Al-Mashhadi
December 4, 2024 at 12:08 PM
Reposted by Zheer Kejlberg Al-Mashhadi
Imagine finding out @vincentab.bsky.social @noahgreifer.bsky.social and @andrew.heiss.phd wrote a thing and it's open access 👀
Awesome paper about {marginaleffects} R/python packages in JOSS (which needs to join Bsky)
www.jstatsoft.org/article/view...
Awesome paper about {marginaleffects} R/python packages in JOSS (which needs to join Bsky)
www.jstatsoft.org/article/view...
How to Interpret Statistical Models Using marginaleffects for R and Python by Vincent Arel-Bundock, Noah Greifer, Andrew Heiss
<p>The parameters of a statistical model can sometimes be difficult to interpret substantively, especially when that model includes nonlinear components, interactions, or transformations. Analysts who...
www.jstatsoft.org
December 1, 2024 at 10:03 PM
Imagine finding out @vincentab.bsky.social @noahgreifer.bsky.social and @andrew.heiss.phd wrote a thing and it's open access 👀
Awesome paper about {marginaleffects} R/python packages in JOSS (which needs to join Bsky)
www.jstatsoft.org/article/view...
Awesome paper about {marginaleffects} R/python packages in JOSS (which needs to join Bsky)
www.jstatsoft.org/article/view...
Reposted by Zheer Kejlberg Al-Mashhadi
I feel this pain
Any good references that states we should not use stepwise regression for causal inference? #causalsky #statsky #episky
November 27, 2024 at 2:40 PM
I feel this pain
Reposted by Zheer Kejlberg Al-Mashhadi
For those who want a quick recap on #epidemiology:
It was plumbing that ended the cholera epidemic, not drinking sewage to strengthen everyone's immune system.
Thanks for coming to my #tedtalk.
It was plumbing that ended the cholera epidemic, not drinking sewage to strengthen everyone's immune system.
Thanks for coming to my #tedtalk.
November 14, 2024 at 10:15 PM
For those who want a quick recap on #epidemiology:
It was plumbing that ended the cholera epidemic, not drinking sewage to strengthen everyone's immune system.
Thanks for coming to my #tedtalk.
It was plumbing that ended the cholera epidemic, not drinking sewage to strengthen everyone's immune system.
Thanks for coming to my #tedtalk.
It’s crazy what causal inference can do now
November 21, 2024 at 2:33 PM
It’s crazy what causal inference can do now
Reposted by Zheer Kejlberg Al-Mashhadi
More helpful (at least for me) resources from the blog archive: Seven (7!) tidyexplain-esque animations (downloadable and CC-licensed) showing how {dplyr}'s mutate(), summarize(), group_by(), and ungroup() all work together #rstats #dataskyence www.andrewheiss.com/blog/2024/04...
November 18, 2024 at 5:26 PM
More helpful (at least for me) resources from the blog archive: Seven (7!) tidyexplain-esque animations (downloadable and CC-licensed) showing how {dplyr}'s mutate(), summarize(), group_by(), and ungroup() all work together #rstats #dataskyence www.andrewheiss.com/blog/2024/04...
Reposted by Zheer Kejlberg Al-Mashhadi
You missed this one!
November 18, 2024 at 8:29 AM
You missed this one!
Just an absolutely fantastic package, the development of which ought to automatically activate some sort of NIH funding
#Rstats `MatchIt` v4.6.0 is out! `MatchIt` implements propensity score matching and other matching methods for causal effect estimation. This isn't a major release, but here are the main updates: 🧵
#causalsky #econsky #episky #statsky
#causalsky #econsky #episky #statsky
November 17, 2024 at 7:49 AM
Just an absolutely fantastic package, the development of which ought to automatically activate some sort of NIH funding
Reposted by Zheer Kejlberg Al-Mashhadi
In clinical research, you will often receive feedback on study design, stats, and/or data analysis from an editor or reviewer that is simply wrong. Here is a list of common "statistical myths" and references you can use to push back.
discourse.datamethods.org/t/reference-...
discourse.datamethods.org/t/reference-...
Reference Collection to push back against "Common Statistical Myths"
Note: This topic is a wiki, meaning that this main body of the topic can be edited by others. Use the Reply button only to post questions or comments about material contained in the body, or to sugge...
discourse.datamethods.org
November 12, 2024 at 6:19 AM
In clinical research, you will often receive feedback on study design, stats, and/or data analysis from an editor or reviewer that is simply wrong. Here is a list of common "statistical myths" and references you can use to push back.
discourse.datamethods.org/t/reference-...
discourse.datamethods.org/t/reference-...
The mood on Bluesky feels like the ending of those zombie apocalypse movies where the protagonists finally find the safe haven.
Just before the sequel where someone on the inside betrays them.
Just before the sequel where someone on the inside betrays them.
November 11, 2024 at 6:12 PM
The mood on Bluesky feels like the ending of those zombie apocalypse movies where the protagonists finally find the safe haven.
Just before the sequel where someone on the inside betrays them.
Just before the sequel where someone on the inside betrays them.