Ed Merkle
LightNews LightNews
banner
edgarmerkle.bsky.social
Ed Merkle
@edgarmerkle.bsky.social
650 followers 120 following 30 posts
quantitative psychologist, open source enthusiast, recently working on R package blavaan
Posts Replies Media Videos
Reposted by Ed Merkle
jonathantemplin.com
A new semester is upon us! This time, I’m teaching Structural Equation Modeling. Want to follow along? Here you go: jonathantemplin.github.io/Structural-E... and www.youtube.com/playlist?lis... #StatModeling #SEM
Structural Equation Modeling (Fall 2025; University of Iowa: PSQF 6249) – Structural Equation Modeling Fall 2025 (PSQF 6249)
jonathantemplin.github.io
August 28, 2025 at 4:29 PM
Reposted by Ed Merkle
pmetricsoc.bsky.social
Don’t miss out on short courses at IMPS 2025! We have 4 short courses to choose from.
www.psychometricsociety.org/imps2025-sho...
#IMPS2025 #Psychometrics #QuantitativePsychology #rstats
Short Courses for IMPS 2025 - Psychometric Society
www.psychometricsociety.org
May 14, 2025 at 3:01 PM
Reposted by Ed Merkle
brandmaier.bsky.social
You want to fit structural equation models in Julia? No problem with the StructuralEquationModels.jl package. Ships with worked examples and tutorials on how to modify the objective function easily (e.g., adding a regularization penalty). Find our preprint here osf.io/preprints/ps...
May 7, 2025 at 1:37 PM
Reposted by Ed Merkle
vincentab.bsky.social
📚😅🎉

Yay!! I just submitted the complete manuscript of my upcoming book to the publisher!

Learn to easily and clearly interpret (almost) any stats model w/ R or Python. Simple ideas, consistent workflow, powerful tools, detailed case studies.

Read it for free @ marginaleffects.com

#RStats #PyData
Model to Meaning: How to interpret statistical models with marginaleffects for R and Python
April 10, 2025 at 7:06 PM
edgarmerkle.bsky.social
Looks like an interesting paper:
Almost Bayesian: The Fractal Dynamics of Stochastic Gradient Descent
We show that the behavior of stochastic gradient descent is related to Bayesian statistics by showing that SGD is effectively diffusion on a fractal landscape, where the fractal dimension can be accou...
doi.org
April 3, 2025 at 3:37 PM
Reposted by Ed Merkle
avehtari.bsky.social
CmdStanR v 0.9.0 released! According to Andrew Johnson, this should make Windows installation easier plus many other improvements and bug fixes discourse.mc-stan.org/t/cmdstanr-v...

#Bayesian #MCMC
CmdStanR: v0.9.0 Release
Hello all, I’m excited to announce that we have now released CmdStanR v0.9.0! This is a minor release, wrapping up several months of bugfixes, quality-of-life improvements, and updates to documentatio...
discourse.mc-stan.org
March 31, 2025 at 10:00 AM
Reposted by Ed Merkle
coatless.bsky.social
Huzzah! Great news for psychometric software!

Psychometrika now accepts software submissions under "Applications and Case Studies" AND software reviews in their expanded Review Section. This dual recognition elevates the status of programming in the field.

#psychometrics #OpenScience
pmetricsoc.bsky.social Psychometric Society @pmetricsoc.bsky.social · Mar 12
New updates to Psychometrika!

- Theory & Methods now also accepts focus articles with discussion/rejoinder

- Book Review is now Review Section, and includes more topics (lit, book, software, media)

- ARCS is now “Applications and Case Studies,” and includes software as a category
March 12, 2025 at 9:07 PM
edgarmerkle.bsky.social
Here is a new post providing some intuition about the proportional odds restriction in ordinal regression models. Maybe you, or your LLM, will find it useful.
The proportional odds restriction – Ed Merkle
Ordinal regression models often employ a proportional odds restriction, which roughly means that the effect of moving from one ordinal category to the next is constant. Here, we provide an example of ...
ecmerkle.github.io
February 21, 2025 at 3:15 PM
Reposted by Ed Merkle
spinkney.bsky.social
Fellow probabilistic programmers I came up with a parameterization for a Cholesky factor of correlation matrices that looks good. I'm asking to test and see if it works well for you discourse.mc-stan.org/t/updated-ch...

#rstats #statistics #bayesian
Updated cholesky corr parameterization testing
Hi everyone, I’m asking if people could test out this parameterization of the Cholesky factor of correlation matrices? From my tests on a macpro m1, it looks to be much faster and I can easily sample...
discourse.mc-stan.org
February 20, 2025 at 1:50 PM
Reposted by Ed Merkle
vincentab.bsky.social
{tinyplot} 0.3.0 is out! 🚨

It's a lightweight #Rstats 📦 to draw beautiful and complex plots, using an ultra-simple and concise syntax.

This is a massive release! @gmcd.bsky.social @zeileis.org and I worked hard to add tons of new themes and plot types.

Check it out!

grantmcdermott.com/tinyplot/
February 5, 2025 at 9:48 PM
Reposted by Ed Merkle
avehtari.bsky.social
If you know simulation based calibration checking (SBC), you will enjoy our new paper "Posterior SBC: Simulation-Based Calibration Checking Conditional on Data" with Teemu Säilynoja, @marvinschmitt.com and @paulbuerkner.com
arxiv.org/abs/2502.03279 1/7
Title: Posterior SBC: Simulation-Based Calibration Checking Conditional on Data Authors: Teemu Säilynoja, Marvin Schmitt, Paul Bürkner, Aki Vehtari Abstract: Simulation-based calibration checking (SBC) refers to the validation of an inference algorithm and model implementation through repeated inference on data simulated from a generative model. In the original and commonly used approach, the generative model uses parameters drawn from the prior, and thus the approach is testing whether the inference works for simulated data generated with parameter values plausible under that prior. This approach is natural and desirable when we want to test whether the inference works for a wide range of datasets we might observe. However, after observing data, we are interested in answering whether the inference works conditional on that particular data. In this paper, we propose posterior SBC and demonstrate how it can be used to validate the inference conditionally on observed data. We illustrate the utility of posterior SBC in three case studies: (1) A simple multilevel model; (2) a model that is governed by differential equations; and (3) a joint integrative neuroscience model which is approximated via amortized Bayesian inference with neural networks.
February 6, 2025 at 10:11 AM
Reposted by Ed Merkle
pmetricsoc.bsky.social
The Psychometric Society invites IMPS 2024 presenters to submit manuscripts to the 2024 Proceedings.

Intent to Submit Deadline: February 14, 2025
Manuscript Deadline: March 21, 2025

www.psychometricsociety.org/imps2024-pro...

#psychometricsociety #psychometrics #quantitativepsychology
January 30, 2025 at 2:56 PM
Reposted by Ed Merkle
adrianhill.de
You think Jacobian and Hessian matrices are prohibitively expensive to compute on your problem? Our latest preprint with @gdalle.bsky.social might change your mind!
arxiv.org/abs/2501.17737
🧵1/8
Figure comparing automatic differentiation (AD) and automatic sparse differentiation (ASD). (a) Given a function f, AD backends return a function computing vector-Jacobian products (VJPs). (b) Standard AD computes Jacobians row-by-row by evaluating VJPs with all standard basis vectors. (c) ASD reduces the number of VJP evaluations by first detecting a sparsity pattern of non-zero values, coloring orthogonal rows in the pattern and simultaneously evaluating VJPs of orthogonal rows. The concepts shown in this figure directly translate to forward-mode, which computes Jacobians column-by-column instead of row-by-row.
January 30, 2025 at 2:32 PM
Reposted by Ed Merkle
clintin.bsky.social
Now out in PNAS, we provide a conceptual framework for assessing replication decisions and replication reform.

Do you view the scientific literature as a "book of truths" or a "book of conversations"?

www.pnas.org/doi/abs/10.1...

#philsci #science #SciSci #Metascience
PNAS
Proceedings of the National Academy of Sciences (PNAS), a peer reviewed journal of the National Academy of Sciences (NAS) - an authoritative source of high-impact, original research that broadly spans...
www.pnas.org
January 28, 2025 at 2:44 PM
Reposted by Ed Merkle
pmetricsoc.bsky.social
Call for Abstracts for IMPS 2025

The Psychometric Society Invites You to Submit an Abstract for IMPS 2025!

The abstract submission portal is scheduled to open Jan 29, 2025

Please see the society website for details: www.psychometricsociety.org/post/imps-20...
IMPS 2025 Abstract Submission - Psychometric Society
Submission Deadline: February 28, 2025
www.psychometricsociety.org
January 22, 2025 at 2:31 PM
Reposted by Ed Merkle
zeileis.org
Our friend and colleague Fritz Leisch died in April last year. In a new contribution to The R Journal we honor Fritz and commemorate his many contributions to science in general and to the R community in particular. #rstats

journal.r-project.org/articles/RJ-...
Remembering Friedrich
This article remembers our friend and colleague Fritz Leisch (1968--2024) who sadly died earlier this year. Many of the readers of The R Journal will know Fritz as a member of the R Core Team and for ...
journal.R-project.org
January 22, 2025 at 3:52 PM
Reposted by Ed Merkle
I am so proud of the Psychometric Society having the courage to transition Psychometrika to Open Access. Practically all APCs are covered by TAs (see bit.ly/3NiAy6T) and for authors not covered will be waived! Support open science for everybody, send your best work to Psychometrika at bit.ly/4eXTvYp
October 9, 2024 at 8:24 PM
edgarmerkle.bsky.social
New preprint with @winterstat.bsky.social and Ellen Fitzsimmons on ordinal factor analysis, where the latent variables are scaled so that they generally take values from 1 to # of ordered categories. The scaling is intuitive because it is what you use when you treat ordinal variables as continuous.
Identification and Scaling of Latent Variables in Ordinal Factor Analysis
Social science researchers are generally accustomed to treating ordinal variables as though they are continuous. In this paper, we consider how identification constraints in ordinal factor analysis ca...
arxiv.org
January 15, 2025 at 2:51 PM
Reposted by Ed Merkle
jebyrnes.bsky.social
OK, here is a very rough draft of a tutorial for #Bayesian #SEM using #brms for #rstats. It needs work, polish, has a lot of questions in it, and I need to add a references section. But, I think a lot of folk will find this useful, so.... jebyrnes.github.io/bayesian_sem... (use issues for comments!)
Full Luxury Bayesian Structural Equation Modeling with brms
jebyrnes.github.io
December 21, 2024 at 7:49 PM
Reposted by Ed Merkle
dsquintana.bsky.social
2. Frontiers for Young Minds

Now this is an excellent initiative. If you haven't heard of this, researchers write papers targeted at kids or teenagers. But the kicker is that your target audience co-peer reviews the paper! Usually, this is done collectively as a science class
Frontiers for Young Minds
Frontiers for Young Minds is an open-access scientific journal written by scientists and reviewed by a board of kids and teens.
kids.frontiersin.org
December 17, 2024 at 9:23 AM
Reposted by Ed Merkle
spmontecarlo.bsky.social
working extra hard on more administrative tasks so that I can clear time to properly read this - looks very interesting!!

arxiv.org/abs/2412.07999
'Fast Mixing of Data Augmentation Algorithms: Bayesian Probit, Logit, and Lasso Regression'
- Holden Lee, Kexin Zhang
December 12, 2024 at 3:25 PM
Reposted by Ed Merkle
njtierney.aus.social.ap.brid.gy
Super stoked to announce that {greta} version 0.5.0 is now on CRAN! Thank you to Nick Golding for making this release happen, and for his continued support, I learnt a lot!

Some features I'm really excited about:

- Now uses Tensorflow 2.0 (which involved a heavy refactor of internals)
- […]
Original post on aus.social
aus.social
November 13, 2024 at 12:31 AM
Reposted by Ed Merkle
cameron.pfiffer.org
Deck.blue is good, you should check it out
November 12, 2024 at 8:38 PM
Reposted by Ed Merkle
jamesuanhoro.bsky.social
Jorgensen & Garnier-Villarreal have a helpful paper showing several failures with "small-variance priors" in Bayesian SEM doi.org/10.1007/978-...

These priors will fail for the types of misspecification we create in simulation studies, wrote a blogpost: www.jamesuanhoro.com/post/2024/11...
The limitations of small variance priors lie in their implementation | James Uanhoro
www.jamesuanhoro.com
November 9, 2024 at 6:41 AM
Reposted by Ed Merkle
colcarroll.bsky.social
I recently got to work on a fun chapter, and the preprint is now available: arxiv.org/abs/2411.04260

We provide intuitions, examples, and recommendations for running MCMC in a world where GPUs and autodiff exist (like this world!)
Running Markov Chain Monte Carlo on Modern Hardware and Software
Today, cheap numerical hardware offers huge amounts of parallel computing power, much of which is used for the task of fitting neural networks to data. Adoption of this hardware to accelerate statisti...
arxiv.org
November 8, 2024 at 3:20 AM