iiinova — where your digital self keeps evolving

Nathaniel Haines @natehaines.bsky.social · 27/08/2026

I'm hiring! My team maintains Oscar's clinical risk adjustment model, which informs everything from member experience to plan pricing, financial reporting, and more. Two roles: - DS 1: www.hioscar.com/careers/7592... - Senior DS: www.hioscar.com/careers/8083...

hioscar.com

Oscar | Smart, simple health insurance.

042

Nathaniel Haines @natehaines.bsky.social · 25/06/2026

i prefer the term "podcast intellectuals" all gotcha hot takes, no technical depth

090

Reposted by Nathaniel Haines

Aki Vehtari @avehtari.bsky.social · 23/06/2026

New paper "To select or not to select: predictively consistent priors instead of model selection" with Anna Elisabeth Riha, Leevi Lindgren, @davidkohns.bsky.social, @paulbuerkner.com arxiv.org/abs/2606.22850 Model selection is not a substitute for building good models in the first place 1/

To select or not to select: predictively
consistent priors instead of model selection

Anna Elisabeth Riha, Leevi Lindgren, David Kohns, Paul-Christian Bürkner, Aki Vehtari

Bayesian modelling workflows often consider multiple candidate models of varying complexity. Model selection is commonly used to navigate potential trade-offs between model complexity and generalisability to new data. We study when model selection is unnecessary or can even be harmful for predictive performance in finite data regimes and find that the need for selecting simpler models can depend on prior choice. We formalise predictively consistent priors, which keep prior predictive implications stable as model complexity increases. Across examples and numerical experiments, including adding covariates in linear and logistic regression, forward variable selection, and nonlinear modelling, flexible models with predictively consistent priors typically match or outperform selected simpler models in out-of-sample predictive performance. When selection helps, it can indicate poor joint prior implications, such as excessive prior mass on implausible predictive values. Based on our findings, we propose replacing the notion of sparsity or parsimony at the level of model components with specifying priors that remain sensible in predictive space as models become more complex.

110631

Reposted by Nathaniel Haines

Henrik Singmann @singmann.bsky.social · 16/06/2026

Finally out in Psychological Review (psycnet.apa.org/doi/10.1037/...), our update to Signal Detection Theory. We show that contrary to the prevailing Gaussian assumption, evidence distributions in recognition memory are likely minimum extreme Gumbel!

Figure 7: Illustration of the Gumbel-min Signal Detection Model
Figure consists of three panels.
Bottom-left panel: The Gumbel-min latent-strength distributions associated with SIGNAL (old) and NOISE (new) items. Top-left panel: The log likelihood ratio (log-LR) for the two latent-strength distributions. Right panel: The receiver operating characteristic function produced by the two latent-strength distributions. pH = hit probabilities; pFA = false-alarm probabilities.

211842

Reposted by Nathaniel Haines

Ricardo Rey-Sáez (β) @ricardoreysaez.bsky.social · 15/06/2026

Finally, our paper is out! Together with @aliciafrancomnez.bsky.social, Javier Revuelta, and @mavadillo.bsky.social, we extend hierarchical factor models for skewed RTs and task validation, now in Psychological Methods: doi.org/10.1037/met0... Let me briefly explain what we did and why it matters!

doi.org

1168

Reposted by Nathaniel Haines

Jonas Dora @jonasdora.bsky.social · 14/06/2026

I'm really excited to share a new preprint! Lab studies have consistently shown that stress and negative emotions make people drink alcohol, but studies of daily life (EMA) keep failing to find it. What does that mean for the shared theoretical prediction both designs test? osf.io/preprints/ps...

Figure summarizing new EMA data from our lab

38226

Reposted by Nathaniel Haines

Fij @mrfijiwiji.bsky.social · 01/06/2026

Historically the word "salad" simply refers to a mixture of chopped cold foods combined with a dressing Therefore, a bowl of cereal with milk is a salad. I call it wheat salad with a milk dressing.

172

Nathaniel Haines @natehaines.bsky.social · 12/06/2026

i have a fun one cooking, stay tuned this next week 🤓

1120

Reposted by Nathaniel Haines

Aki Vehtari @avehtari.bsky.social · 08/06/2026

Two recommendations for cmdstanr in case studies: 1) use sample(..., refresh=0) to avoid 80 lines of MCMC progress reporting 2) cmdstanr::print_stan_file() in Quarto with output asis to get color formatting ``` #| output: asis print_stan_file(stan_file) ``` e.g., avehtari.github.io/Bayesian-Wor...

$print_stan_file(bioassay_stan_file) data { int<lower=0> J; vector[J] x; array[J] int<lower=0> N; array[J] int<lower=0, upper=N> y; } parameters { real a; real<lower=0> b; } model { {a, b} ~ normal(0, 5); y ~ binomial_logit(N, a + b * x); }$

1346

Reposted by Nathaniel Haines

Thùy Vy T Nguyễn @thuyvytnguyen.bsky.social · 17/02/2026

Happy New Year! It is the Year of the Fire Horse. I want to use this turn of the year to share that as of today, I have stepped down from my position as Associate Professor at @durhampsych.bsky.social and will now serve full time as director of my non-profit, @thesolitudelab.bsky.social,...

3183

Nathaniel Haines @natehaines.bsky.social · 03/06/2026

giving a talk later this year and this is gonna be so fun

0116

Nathaniel Haines @natehaines.bsky.social · 28/05/2026

the "methods" section is the only thing that separates science from journalism

120

Nathaniel Haines @natehaines.bsky.social · 24/05/2026

reach out if you or someone you knows wants to chat about doing some modeling together 🤓 my focus is primarily on modeling behavioral paradigms and self-report measures—either together, separately, or even longitudinally bayesianbeginnings.com/?v=2

bayesianbeginnings.com

Bayesian Beginnings — Consulting for Bayesian & Computational Modeling

Helping researchers design, fit, and validate Bayesian and computational models — and build the reproducible pipelines that make the work last.

0144

Nathaniel Haines @natehaines.bsky.social · 20/05/2026

some big news! hBayesDM has now been fully ported to cmdstan/cmdstanr/cmdstanpy check out the PR below, or the changelog for an overview: PR: github.com/CCS-Lab/hBay... Changelog: ccs-lab.github.io/hBayesDM/new... stay tuned for some additional features coming soon, including covariate support 🤓

github.com

feature: hBayesDM version 2.0 by Nathaniel-Haines · Pull Request #182 · CCS-Lab/hBayesDM

hBayesDM 2.0 — modernize stack & toolchain Top-to-bottom refactor of both the R and Python packages onto current Stan tooling, plus the supporting work to keep behavior, docs, and CI in sync. H...

2154

Nathaniel Haines @natehaines.bsky.social · 18/05/2026

i like the approach here tbh it's rather extreme, but warranted if you can't be bothered to do some basic due diligence on papers whose purpose is to spread knowledge there are plenty of forums for non fact checked, sloppy writing anyway

140

Nathaniel Haines @natehaines.bsky.social · 16/05/2026

same but influencer -> psych researcher 4 humors -> big 5

020

Nathaniel Haines @natehaines.bsky.social · 15/05/2026

even if llms are only good at local interpolation, i think it's easy to lose sight of how vast (and sparse) the sample space is, and therefore how useful even minor interpolation can be

030

Reposted by Nathaniel Haines

Riccardo Fusaroli @fusaroli.eurosky.social · 04/05/2026

some Bayesian cognitive modeling students are having trouble seeing the industry-relevance of the methods (my worked out examples are research based, since it's what I do). Tips for good industry examples? @ajordannafa.com @natehaines.bsky.social @rmcelreath.bsky.social @avehtari.bsky.social

72114

Reposted by Nathaniel Haines

Berna Devezer @devezer.bsky.social · 30/04/2026

📣 The difference between replicable and not replicable is not itself scientifically replicable. 📣 New work with Erkan Buzbas, showing that verdicts such as "X% of results replicated" are based on an inferential machinery that doesn't work. arxiv.org/abs/2604.26268

Screenshot of a paper's title page. Title: "The Difference Between 'Replicable' and 'Not replicable' is not Itself Scientifically Replicable". Authors: Berna Devezer and Erkan O. Buzbas, both at the University of Idaho — Devezer in the Department of Business and the Institute for Modeling Collaboration and Innovation, Buzbas in the Department of Mathematics and Statistical Science. Authors contributed equally. Corresponding author: bdevezer@uidaho.edu.
Abstract: Replication studies estimate the replicability rate of scientific results by aggregating binary verdicts of experiments. Exact replications are rarely attainable, so most replication sequences are non-exact. Experiments differ in ways that matter and do not share a single common data-generating process. We formalize two statistical interpretations of this non-exactness. In a shared latent rate model (benchmark), experiments are exchangeable and depend on a common random replicability rate. In a conditionally independent rates model (operational), each experiment has its own replicability rate drawn independently from a population distribution. Under the shared latent rate model, even small variability among replicability rates induces an irreducible variance floor on the estimated mean replicability rate that cannot be eliminated by adding more replications. Under the conditionally independent rates model, the degree of non-exactness is not identifiable from standard replication data, because one binary verdict per experiment contains no information about between-experiment heterogeneity. Researchers therefore cannot tell which precision regime they are operating in or whether high- and low-replicability sequences can be distinguished in principle. As a result, the usual data structure of one binary verdict per experiment cannot support reliable demarcation between "replicable" and "not replicable" results and systematically understates uncertainty, making high- and low-replicability sequences appear discrim…

15547205

Nathaniel Haines @natehaines.bsky.social · 24/04/2026

disqus just started lighting up my blog comment section with ads unfortunate probably going to just delete the comment section this weekend 😬

000

Nathaniel Haines @natehaines.bsky.social · 02/04/2026

TBF this is a trend that has been going on for quite some time. Statistical software has become so user friendly that you can do some rather complex analyses these days without truly understanding your model Is this a net bad? Hard to say. It’s on users to do their due diligence IMO

1214

Nathaniel Haines @natehaines.bsky.social · 01/04/2026

one day I will convince folks to actually use their reliability estimates and not simply report them... one day...

161

Reposted by Nathaniel Haines

Kevin O’Neill @kevingoneill.github.io · 31/03/2026

it’s hard to overstate how important this is for quantitative research (especially psychology/neuroscience)- computing individual-level stats independently often robs you of hard earned statistical power. highly recommended reading *both* of these wonderful blog posts!

2194

Reposted by Nathaniel Haines

Nathaniel Haines @natehaines.bsky.social · 31/03/2026

And in what is my favorite figure ever, we look into how shrinkage impacts person-level means in both the univariate and multivariate case:

272

Nathaniel Haines @natehaines.bsky.social · 31/03/2026

Part 2 of my shrinkage estimator series is out! Part 1 covered the univariate case, but now we dive into multivariate shrinkage 🤓 We cover Spearman's classic correlation disattenuation formula, multivariate James-Stein estimators, and hierarchical methods too haines-lab.com/post/how-to-...

haines-lab.com

How to Estimate a Correlation, and What It Means for Science | Computational Psychology

Introduction In Part 1 of this series, we showed that six different methods for estimating individual means—James-Stein, classical true score estimation, empirical Bayes, ridge regression, and hierarc...

24215

Nathaniel Haines @natehaines.bsky.social · 26/03/2026

guys i got a review request for a paper exploring if LLMs can instantiate "computations of psychopathology" wut

010

Reposted by Nathaniel Haines

Lukas A. Basedow, PhD @labasedow.bsky.social · 21/03/2026

One of my main scientific regrets is not having used my time to learn more statistics. Learning new statistical approaches next to general postdoc work is not very sustainable. However, fantastic blog posts such as this one by @natehaines.bsky.social really help! haines-lab.com/post/how-to-...

haines-lab.com

How to Estimate a Mean, and What It Means for Science | Computational Psychology

Introduction If I asked you to estimate 30 means, you would probably compute 30 sample means. And you would be provably wrong (well.. maybe not wrong, but at least provably inefficient).

161

Nathaniel Haines @natehaines.bsky.social · 17/03/2026

good bayesians don't let their friends do NHST with bayes factors

1166

Nathaniel Haines @natehaines.bsky.social · 16/03/2026

my favorite part of this table is the implication that the best estimators are essentially those that leverage priors frequentists try really hard to avoid the concept of a prior, and yet... if you have ever fit a hierarchical model you are practically Bayesian 😈

150

Nathaniel Haines @natehaines.bsky.social · 15/03/2026

New blog just dropped! This one is all about estimators—we cover James-Stein, classical test theory, empirical Bayes, penalized regression, and hierarchical models, showing how they all can be used to do a better job than sample stats alone 🤓 haines-lab.com/post/how-to-...

haines-lab.com

How to Estimate a Mean, and What It Means for Science | Computational Psychology

Introduction If I asked you to estimate 30 means, you would probably compute 30 sample means. And you would be provably wrong (well.. maybe not wrong, but at least provably inefficient).

56524

Nathaniel Haines @natehaines.bsky.social · 30/01/2026

An aside—LLMs have basically solved data viz for me. What used to take a good bit of tinkering in ggplot or plotly now can be one-shotted with a data snippet/schema as context and a plain English description of what I want. The bar for data viz is now higher 🤓

0110

Nathaniel Haines @natehaines.bsky.social · 26/01/2026

1/9 New blog is live! This is part 2 of a series—last time we looked at the Dunning-Kruger effect, now we are digging in to Implicit vs Explicit attitudes and the Implicit Association Test. To start, of course we need a good meme... haines-lab.com/post/part-2-...

44214

Nathaniel Haines @natehaines.bsky.social · 19/01/2026

coming soon! part 2 of the series (finally..), this time looking at implicit attitudes instead of the Dunning-Kruger effect—here is a sneak peek 🤓

290

Nathaniel Haines @natehaines.bsky.social · 14/01/2026

when McElreath drops new videos you watch, and it is always so worth it look at this awesome graph, no better way to illustrate how cool hierarchical Bayes is 😍 youtu.be/jh3RltVrQ-Q?...

0162

Reposted by Nathaniel Haines

Holly Sullivan-Toole @hollysully.bsky.social · 07/01/2026

New paper & a thread on the results 👇 ‘Reward-specific learning parameters change across normative adolescent development and are blunted in youth with high risk for depression’ acamh-onlinelibrary-wiley-com.ezp3.lib.umn.edu/doi/full/10....

screenshot of article title & author team

42311

Nathaniel Haines @natehaines.bsky.social · 14/12/2025

eyyy survived another year! go team, your citations are appreciated 🙏

110

Nathaniel Haines @natehaines.bsky.social · 07/12/2025

cmon guys 2025 doesn't need to be the inflection point, help me out

150

Reposted by Nathaniel Haines

Python Software Foundation @python.org · 27/10/2025

TLDR; The PSF has made the decision to put our community and our shared diversity, equity, and inclusion values ahead of seeking $1.5M in new revenue. Please read and share. pyfound.blogspot.com/2025/10/NSF-... 🧵

python.org

The official home of the Python Programming Language

12363772727

Nathaniel Haines @natehaines.bsky.social · 23/10/2025

just got a restitution check for my auto deductible from back when I got kia boi'd a few years ago.. what a pleasant surprise

000

Nathaniel Haines @natehaines.bsky.social · 16/09/2025

awesome post, and couldn't agree more with the thesis R is great because it "just works" most of the time

030

Nathaniel Haines @natehaines.bsky.social · 19/08/2025

Job update! It has been a little over a month now, but I'm excited to share that I have joined Oscar Health, where I'm leading a team focused on our clinical risk adjustment process 🤓

1110