alex hayes
@alexpghayes.com
postdoc @ stanford econ + incoming assistant prof @ oregon state statistics. networks, causal inference, contagion, measurement error, #rstats. he/him
https://www.alexpghayes.com
https://www.alexpghayes.com
Do the RStudio Check/Test/Load all keybinds work in Positron?
November 6, 2025 at 9:50 PM
Do the RStudio Check/Test/Load all keybinds work in Positron?
there's some recently theory on optimal tests in arxiv.org/abs/2107.07575, with accompanying code in achambaz.github.io/mediation.te...
tl;dr it's a complicated hypothesis to test
tl;dr it's a complicated hypothesis to test
Optimal tests of the composite null hypothesis arising in mediation analysis
The indirect effect of an exposure on an outcome through an intermediate variable can be identified by a product of two regression coefficients under certain causal and regression modeling assumptions...
arxiv.org
November 5, 2025 at 9:02 PM
there's some recently theory on optimal tests in arxiv.org/abs/2107.07575, with accompanying code in achambaz.github.io/mediation.te...
tl;dr it's a complicated hypothesis to test
tl;dr it's a complicated hypothesis to test
Just read this -- really enjoy the differencing out of the calibration curve to find questions that folks are most differentially miscalibrated on
November 5, 2025 at 7:59 PM
Just read this -- really enjoy the differencing out of the calibration curve to find questions that folks are most differentially miscalibrated on
Just ordered a copy! Thanks so much for all these resources :)
November 4, 2025 at 7:27 PM
Just ordered a copy! Thanks so much for all these resources :)
@andrew.heiss.phd pinging you because I suspect you'll have ideas
November 4, 2025 at 1:45 AM
@andrew.heiss.phd pinging you because I suspect you'll have ideas
I also know some applied folks who like pre-registration simply because it forces them to plan an analysis before they run the experiment, and they've caught experimental design errors this way
November 2, 2025 at 5:09 PM
I also know some applied folks who like pre-registration simply because it forces them to plan an analysis before they run the experiment, and they've caught experimental design errors this way
What's the context for the last one?
October 3, 2025 at 6:17 AM
What's the context for the last one?
Some of Andrew Vickers work on Net Benefit might be relevant, although that's written for a clinical context
e.g., journals.sagepub.com/doi/10.1177/...
e.g., journals.sagepub.com/doi/10.1177/...
Calibration of Risk Prediction Models - Ben Van Calster, Andrew J. Vickers, 2015
Decision-analytic measures to assess clinical utility of prediction models and diagnostic tests incorporate the relative clinical consequences of true and false...
journals.sagepub.com
August 28, 2025 at 8:43 PM
Some of Andrew Vickers work on Net Benefit might be relevant, although that's written for a clinical context
e.g., journals.sagepub.com/doi/10.1177/...
e.g., journals.sagepub.com/doi/10.1177/...
Can you share some pointers to this literature? I'm curious to read more
August 25, 2025 at 6:27 AM
Can you share some pointers to this literature? I'm curious to read more
Once again enjoying these materials. Do you know of any typography resources with Latex/Typst/Quarto example code?
Practical Typography is fascinating but I'm unsure how to implement many of the recommendations
Practical Typography is fascinating but I'm unsure how to implement many of the recommendations
August 22, 2025 at 6:45 AM
Once again enjoying these materials. Do you know of any typography resources with Latex/Typst/Quarto example code?
Practical Typography is fascinating but I'm unsure how to implement many of the recommendations
Practical Typography is fascinating but I'm unsure how to implement many of the recommendations
The caveat being that you always get clusters back so you need ways to distinguish between good and bad clusters
August 13, 2025 at 2:10 PM
The caveat being that you always get clusters back so you need ways to distinguish between good and bad clusters
I think the value of clustering is more often about EDA and discovery rather than confirmatory work or solving a known problem
August 13, 2025 at 2:10 PM
I think the value of clustering is more often about EDA and discovery rather than confirmatory work or solving a known problem
Reposted by alex hayes
Also, we think people should stop using forensic meta-analytic techniques that have not been adequately vetted by experts in statistics. The p curve papers have very little statistical detail, and were published in psych journals. They did not get scrutiny appropriate to their popularity. 15/?
August 8, 2025 at 6:56 PM
Also, we think people should stop using forensic meta-analytic techniques that have not been adequately vetted by experts in statistics. The p curve papers have very little statistical detail, and were published in psych journals. They did not get scrutiny appropriate to their popularity. 15/?
Is there an alternative that does these things?
August 9, 2025 at 9:18 PM
Is there an alternative that does these things?