Andrei

Wouldn’t using a class S7 object be more efficient and less error prone?

Compositional data (proportions that sum to 1) behave in ways standard models aren’t built for

I walk through why Dirichlet regression is often the right tool & what extra insight it gives using a real ex of eyetracking

#Dirichlet #r #brms #guide #eyetracking

open.substack.com/pub/mzlotean...

One approach to the age-period-cohort problem: Just don’t. Just to cause yourself more problems, you seek for something. But there is no need for you to seek anything. You have plenty, and you have just enough problems. Shunryū Suzuki in a 1971 talk A ...

New blog post about the age-period-cohort identification problem!

In which, for the first time ever, I ask "What's the mechanism?" and also suggest that sometimes you may actually *not* be interested in causal inference.

www.the100.ci/2026/02/13/o...

Results for CEM and neighborhood fixed effects regressions. We find robust and significant positive effects (blue coefficients) of childhood exposure to different ethnicities on the likelihood of interethnic marriage.

Left: georeferenced households in 1880. Right: Ethnic organic neighborhoods in Manhattan in 1880. Six ethnic groups are prevalent: first/second generation Americans, Asians, Germans, Irish and Others (residual category). We use these neighborhoods to account for segregation by applying organic neighborhood fixed effects.

New preprint out with @wendering.bsky.social & Nan Zhang!

We show that childhood exposure to ethnic outgroups increases the prob. of #interethnic marriage decades later, using historical linked US census data (1880–1910) and next-door neighbor comparisons 🏠🌃

Read more here:
shorturl.at/U0IHR

The joke of the week in my causal inference methods course this week 😂

Opinion | Students Are Skipping the Hardest Part of Growing Up

We ran a massive, uncontrolled social experiment on kids with social media. Outcomes: anxiety, comparison, fractured attention. Jon Haidt called the direction of travel. We’re doing it again with AI, outsourcing thinking and social skills, to act surprised later. www.nytimes.com/2026/01/30/o...

Learn linux, that's it...

Bayesian model comparison implements Occam's razor through its sensitivity to the prior. However, prior-dependence makes it important to assess the influence of plausible alternative priors. Such prior sensitivity analyses for the Bayesian evidence are expensive, either requiring repeated, costly model re-fits or specialised sampling schemes. By exploiting the learned harmonic mean estimator (LHME) for evidence calculation we decouple sampling and evidence calculation, allowing resampled posterior draws to be used directly to calculate the evidence without further likelihood evaluations. This provides an alternative approach to prior sensitivity analysis for Bayesian model comparison that dramatically alleviates the computational cost and is agnostic to the method used to generate posterior samples. We validate our method on toy problems and a cosmological case study, reproducing estimates obtained by full Markov chain Monte Carlo (MCMC) sampling and nested sampling re-fits. For the cosmological example considered our approach achieves up to $6000\times$ lower computational cost.

arXiv📈🤖
Efficient prior sensitivity analysis for Bayesian model comparison
By Hu, McEwen

Finally got around to removing broom::tidy(), broom::glance(), and broom::augment() from my class examples in favor of parameters::model_parameters(), performance::model_performance() and marginaleffects::predictions() because they're *so nice* for teaching! #rstats #easystats

Agent Psychosis: Are We Going Insane? What’s going on with the AI builder community right now?

Great cogent post on the issues of agentic coding (Eg our buddy Claude Code) and the spillover costs on everyone else

lucumr.pocoo.org/2026/1/18/ag...

@gmcd.bsky.social hits on the things you were complaining about

The way I understand Trumpism is similar to the rise of fascism in the 30s:

There was a wave of globalization and liberalization of trade and migration which had clear winners and losers. When the working class lost faith in the system, capitalists aligned with ethno-nationalists to preserve it.

I am thrilled that my book, "Claiming the Right to the City: Rethinking Urban Transformations in Brazil," is now published at @ubcpress.bsky.social. It explores profound divisions between the right to the city on paper and the reality in practice. See bit.ly/49v96gQ for more info, or get in touch.

New AI agentic research agent. Claims to procide accelerated scientific discovery pipeline by:

- 🧠 Research Co-Pilot Intelligence
- ⚙️ Autonomous Algorithm Innovation
- 📊 Intelligent Data Orchestration
- 🔬 Scientific Reproducibility Engine
- 📚 AI-Powered Deep Survey

See for yourself: novix.science

«Sin vivienda social no puede haber proximidad en las ciudades» Carlos Moreno  Urbanista

Mi entrevista en La Razón de España
La vivienda social es indispensable tenerla en toda la ciudad para generar mixtura social, luchar contra la gentrificación, acceder a servicios, calidad de vida y promover una proximidad feliz.
Alojarse no es vivir en la ciudad!
www.larazon.es/medio-ambien...

Contextual Distraction: RAG isn't a Seatbelt A laboratory safety benchmark finds retrieval augmented generation (RAG) can make strong models worse.

Contextual Distraction: RAG isn't a Seatbelt. A laboratory safety benchmark finds retrieval augmented generation (RAG) can make strong models worse. doi.org/10.59350/8v4... #biosecurity 🧬🖥️🧪

The beauty of academia? Total freedom to pick which 7 days you work each week.

Count-compositional data arise in many different fields, including high-throughput microbiome sequencing and palynology experiments, where a common, important goal is to understand how covariates relate to the observed compositions. Existing methods often fail to simultaneously address key challenges inherent in such data, namely: overdispersion, an excess of zeros, cross-sample heterogeneity, and nonlinear covariate effects. To address these concerns, we propose novel Bayesian models based on ensembles of regression trees. Specifically, we leverage the recently introduced zero-and-$N$-inflated multinomial distribution and assign independent nonparametric Bayesian additive regression tree (BART) priors to both the compositional and structural zero probability components of our model, to flexibly capture covariate effects. We further extend this by adding latent random effects to capture overdispersion and more general dependence structures among the categories. We develop an efficient inferential algorithm combining recent data augmentation schemes with established BART sampling routines. We evaluate our proposed models in simulation studies and illustrate their applicability with two case studies in microbiome and palaeoclimate modelling.

arXiv📈🤖
Bayesian nonparametric models for zero-inflated count-compositional data using ensembles of regression trees
By Menezes, Parnell, Murphy

💯

I’m a data scientist who is skeptical about data After millennia of relying on anecdotes, instincts, and old wives’ tales as evidence of our opinions, most of us today demand that people use data to support their arguments and ideas. Whether it’s cu...

For the "data scientists" out there who are throwing things at machines (e.g., LLMs) and trying to extract "insights," I recommend (re)reading this piece:

"Data is therefore only as useful as its quality and the skills of the person wielding it."

qz.com/1664575/is-d...

wordcount plug in output with word counts

My #rstats / #python tip of the day is @andrew.heiss.phd wordcount Quarto plug in.

Just change one line in your YAML header and your Quarto doc will produce separate word counts for body, notes, references, etc.

See link here: github.com/andrewheiss/...

Young adult life courses in the Global South: A comparative framework and research agenda As of 2025, 85% of the world population lives in countries typically associated with the Global South. Fifty percent of the world’s population is unde…

🚨New year, new paper with the fab Anette Fasang, Ignacio Cabib, Adam Cooper &
@robgruijters.bsky.social!

We introduce our SI on 'Young Adult Life Courses in the Global South' & develop a new framework and concepts to advance the field. www.sciencedirect.com/science/arti...

@sriucl.bsky.social

arXiv:2307.09404v3 Announce Type: replace Abstract: The starting point for much of multivariate analysis (MVA) is an $n\times p$ data matrix whose $n$ rows represent observations and whose $p$ columns represent variables. Some multivariate data sets, however, may be best conceptualized not as $n$ discrete $p$-variate observations, but as $p$ curves or functions defined on a common time interval. Here we introduce a framework for extending techniques of multivariate analysis to such settings. The proposed continuous-time multivariate analysis (CTMVA) framework rests on the assumption that the curves can be represented as linear combinations of basis functions such as $B$-splines, as in the Ramsay-Silverman representation of functional data; but whereas functional data analysis extends MVA to the case of observations that are curves rather than vectors -- heuristically, $n\times p$ data with $p$ infinite -- we are instead concerned with what happens when $n$ is infinite. We present continuous-time extensions of the classical MVA methods of covariance and correlation estimation, principal component analysis, Fisher's linear discriminant analysis, and $k$-means clustering. We show that CTMVA can improve on the performance of classical MVA, in particular for correlation estimation and clustering, and can be applied in some settings where classical MVA cannot, including variables observed at disparate time points. CTMVA is illustrated with a novel perspective on a well-known Canadian weather data set, and with applications to data sets involving international development, brain signals, and air quality. The proposed methods are implemented in the publicly available R package \texttt{ctmva}.

arXiv📈🤖 New Paper
Continuous-time multivariate analysis
By

Mamdani has only been Mayor for a day & criminals are already fleeing New York

Temperatures have struggled to climb above freezing in places today, with frost returning already

Significantly colder Arctic air is gathering just to our N now though, behind the cold front you can see on the pressure chart and on satellite imagery

This sweeps S next 24hrs

Happy New Year, everyone! 🎉
Because my work focuses on 𝘩𝘰𝘶𝘴𝘪𝘯𝘨 𝘢𝘧𝘧𝘰𝘳𝘥𝘢𝘣𝘪𝘭𝘪𝘵𝘺 and 𝘶𝘳𝘣𝘢𝘯 𝘴𝘶𝘴𝘵𝘢𝘪𝘯𝘢𝘣𝘪𝘭𝘪𝘵𝘺, I put together a page that collects and periodically updates 𝗿𝗲𝗰𝗲𝗻𝘁 𝗲𝗺𝗽𝗶𝗿𝗶𝗰𝗮𝗹 𝘀𝘁𝘂𝗱𝗶𝗲𝘀 𝗼𝗻 𝗿𝗲𝗻𝘁 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 𝗮𝗻𝗱 𝗶𝘁𝘀 𝗲𝗳𝗳𝗲𝗰𝘁𝘀 𝘄𝗼𝗿𝗹𝗱𝘄𝗶𝗱𝗲. 📚🌍
👉 sites.google.com/view/hjiang/...

A nice objective for 2026!

Would add that you can use a local LLM (e.g. AnythingLLM) with a Zotero + Notion MCP connection as a knowledge base. Giving you the ability to query notes and papers.

How to Use Notion for Academic Research – Homepage I provide some advice for how to use Notion’s powerful management software for the unique challenges of academic research.

🚨 New blog post: How to Use Notion for Academic Research

I've used Notion for some time to manage my workflow & collaborate, and I've often had Qs about how I use it. This post summarizes the way I manage tasks & projects using the free academic account.

Link: www.robertkubinec.com/post/notion/

In our final wrap-up of 2025, we explore different facets of inequality around the world – from gender justice in Bangladesh and racial divides in South Africa, to the unfulfilled promises of democracy in Brazil – as well as some “global” perspectives.

#LSEInequalitiesBlog

In Nature >> Science in 2050: the future breakthroughs that will shape our world — and beyond
-
www.nature.com/articles/d41...