John Poe

3K posts

John Poe

John Poe

@DavidPoe223

Methodologist who does multilevel & panel data things. PhD in PoliSci. Likes=bookmarks 🌈

Cabot, AR Katılım Temmuz 2016
1.8K Takip Edilen1.4K Takipçiler
Sabitlenmiş Tweet
John Poe
John Poe@DavidPoe223·
Here's an attempt at posting the brochure for my course offerings this year. If you're interested in Modern DiD, Advanced Multilevel Modeling, and Bayesian Multilevel Modeling then take a look. You can register on my website or @ICPSRSummer
John Poe tweet mediaJohn Poe tweet mediaJohn Poe tweet mediaJohn Poe tweet media
English
1
6
37
5.8K
John Poe retweetledi
Nick HK
Nick HK@nickchk·
I am teaching my Causal Inference in Econometrics asynchronous workshop with Statistical Horizons from April 13-May 11 of this year. Learn about the tools of causal inference and go at your own pace. Get more info, and sign up if you're interested, here: statisticalhorizons.com/seminars/causa…
English
1
3
29
2.2K
John Poe retweetledi
Arpit Gupta
Arpit Gupta@arpitrage·
These results suggest that getting a PhD causally worsens mental health, or at least receiving psychiatric medicines. The reversal post PhD degree is particularly convincing. But the up trend among the control group is intriguing. The highly educated are in distress.
Arpit Gupta tweet mediaArpit Gupta tweet mediaArpit Gupta tweet mediaArpit Gupta tweet media
English
70
371
1.8K
469.5K
John Poe
John Poe@DavidPoe223·
I don't think I can do this again. I can't take another round of this nightmare
English
0
0
2
287
John Poe retweetledi
Melissa S. Kearney
Melissa S. Kearney@kearney_melissa·
Somewhere along the way “correlation is not causation” morphed into “until the causal link is proven beyond a reasonable doubt, ideally with a large scale RCT with global external validity, than the correlation likely reflects the impact of some unobserved factor I can’t name.”
English
88
290
2.3K
383.6K
John Poe
John Poe@DavidPoe223·
@PhDemetri What makes it so interesting to you? I know the basics of it from going down a rabbit hole of shrinkage techniques, but I've never actually used it for anything.
English
1
0
1
986
John Poe
John Poe@DavidPoe223·
I'm trying to write an R package for the first time and (*screams incoherently!*), it's kind of frustrating.
English
6
0
11
1K
John Poe
John Poe@DavidPoe223·
@stephenjwild It's more or less how I write equations with specific examples for course notes. The big difference is I align the + with one object on each line vertically. Looking at this makes my skin crawl
English
1
0
4
550
John Poe retweetledi
Jeffrey Wooldridge
Jeffrey Wooldridge@jmwooldridge·
Is it really true that, for something like Poisson regression with panel data -- say, in a nonlinear diff-in-diffs setting -- R does not allow an option for clustered standard errors within the command? If so, that's a serious shortcoming of R. Big win for Stata.
English
8
18
279
82.3K
John Poe
John Poe@DavidPoe223·
@stephenjwild @rubenarslan GLMs are really fragile to model misspecification to begin with so they'll be more likely to break if you enter don't model a random effect or you get the wrong distribution of the REs from whatever tool you're using to estimate it
English
0
0
1
168
Stephen Wild
Stephen Wild@stephenjwild·
@rubenarslan I am an amateur too, so I thought maybe I had missed something 😅 IIRC, I think for linear models there's no bias, but not for glms. But someone more knowledgeable than me will know
English
1
0
0
207
Ruben C. Arslan
Ruben C. Arslan@rubenarslan·
My student asked whether we need to model the slopes of covariates as varying, if we do so for the slope of our main exposure. I thought not doing so could bias the estimate, but a quick simulation only showed increased variance. Wrong? Let me know. rubenarslan.github.io/posts/2024-05-…
English
5
6
35
9.3K
John Poe
John Poe@DavidPoe223·
@rubenarslan Your intuition is partly right. In cases where you've got Simpsons paradox and don't use group means then the random slope will help debias the main effect some but not totally. Otherwise it's just showing you that there is variation in effects around the average.
English
0
0
4
213
John Poe
John Poe@DavidPoe223·
I'm teaching my Modern DiD class for @ICPSRSummer in a few weeks. Reviewing my notes reminds me how quickly this literature has grown! I think im going to have to add in a lecture just for strategies to keep track of updates
English
1
0
4
401
John Poe
John Poe@DavidPoe223·
I'll agree and add that you have to know the coding language well enough to proofread it carefully. You also need to build checks directly into it.
English
0
0
2
299
John Poe retweetledi
Noah Greifer
Noah Greifer@noah_greifer·
This is a must-read for anyone interested in causal inference methods, especially if TMLE or DML are opaque to you. Well written as always @ildiazm. Your clarity, rigor, and expertise are inspiring.
Lars van der Laan@LarsvanderLaan3

What are the differences between one-step estimation, Double ML, and Targeted ML? This commentary (@ildiazm) and blog post (@mark_vdlaan) provide an overview of the history of machine learning in semiparametrics. pubmed.ncbi.nlm.nih.gov/31742333/ vanderlaan-lab.org/2019/12/24/cv-…

English
1
4
42
6.9K
John Poe retweetledi
Sarah Miller
Sarah Miller@smilleralert·
It's INCREDIBLE that longitudinal admin data on tax records, recidivism, health, and much more, are now publicly and freely available. Wow! So many diff in diffs to run!! Congratulations @UM_CJARS! I am proud to have you as my colleagues!
CJARS@UM_CJARS

Today, we are proud to introduce the #JusticeOutcomesExplorer! JOE is a publicly available data dashboard that provides a fresh glimpse into how the criminal justice system touches the lives of millions of Americans every year: joe.cjars.org [1/10] 🧵

English
2
32
143
26.7K