iiinova — where your digital self keeps evolving

Stats For Bios @statsforbios.bsky.social · 05/10/2026

*disagapp... not shinyapp. eejit.

Stats For Bios @statsforbios.bsky.social · 05/10/2026

New preprint from my student Carolina Molina-Tabares. Different to other preprints I post! She evaluated a portable qPCR workflow for rapid on-site detection of Ae. aegypti under field conditions. Compared to benchtop Roche LightCycler 96. Tested in Kenya. www.researchsquare.com/article/rs-1...

Photo and diagram of the field setup. A portable qPCR machine, run on a motorcycle battery, plugged into a laptop.

000

Stats For Bios @statsforbios.bsky.social · 04/10/2026

Absolutely agree. My only parenting advice ever has been "take all parenting advice with a huge fistful of salt".

020

Reposted by Stats For Bios

Andrew Gelman et al. @statmodeling.bsky.social · 02/10/2026

If you were planning to send me a chatbot-written email or post a chatbot-written comment, just send or post your prompt. That’s enough. You can keep the slop to yourself. statmodeling.stat.columbia.edu/2026/10/02/i...

statmodeling.stat.columbia.edu

If you were planning to send me a chatbot-written email or post a chatbot-written comment, just send or post your prompt. That’s enough. You can keep the slop to yourself. | Statistical Modeling, C...

0243

Reposted by Stats For Bios

Carl T. Bergstrom @carlbergstrom.com · 03/10/2026

I have absolutely wondered about this. Amazing.

311800322

Reposted by Stats For Bios

Henrik Bengtsson @henrikbengtsson.bsky.social · 01/10/2026

Oh, great. TIL #RStats sprintf("%03s", x) differ on macOS and Linux; Linux: > sprintf("x=%02s", c("1", "2", "a")) [1] "x= 1" "x= 2" "x= a" macOS: > sprintf("x=%02s", c("1", "2", "a")) [1] "x=01" "x=02" "x=0a" Personally, I'd argue padding strings with zeros should not be an error. Oh well...

381

Stats For Bios @statsforbios.bsky.social · 29/09/2026

Why does windows default to not showing file extensions? So stupid. And a really nice way to make teaching into R more difficult. #RStats

050

Reposted by Stats For Bios

Rod Page @rdmpage.bsky.social · 23/09/2026

Jobs galore at @nhm-london.bsky.social @dissco.bsky.social - jobs.nhm.ac.uk/Job/JobDetai... - jobs.nhm.ac.uk/Job/JobDetai... - jobs.nhm.ac.uk/Job/JobDetai... - jobs.nhm.ac.uk/Job/JobDetai... - jobs.nhm.ac.uk/Job/JobDetai... including Heads of Digitisation and Technical Innovation

jobs.nhm.ac.uk

DiSSCo UK Digitiser:South Kensington

0412

Reposted by Stats For Bios

Saganism @sagan.bsky.social · 22/09/2026

A humbling reminder from Carl Sagan: Even if you read 1 book a week for your entire life, you'd only scratch the surface of human knowledge.

221137295

Reposted by Stats For Bios

MC Stan @mc-stan.org · 17/09/2026

Release of CmdStan 2.40, including new Stan Math, core Stan, and Stanc, is now available! Blog post with new features blog.mc-stan.org/2026/09/16/r... Highlights - integrate_1d_double_exponential, a variadic version of old integrate_1d - integrate_1d_gauss_kronrod, a different quadrature rule ...

blog.mc-stan.org

Release of CmdStan 2.40

We are very happy to announce that the 2.40.0 release of CmdStan is now available on Github! As usual, the release of CmdStan is accompanied by new releases of Stan Math, core Stan, and Stanc3. Thi…

13016

Stats For Bios @statsforbios.bsky.social · 17/09/2026

Had a great week teaching Geospatial Models for Public Health at Tanzanian National Bureau of Statistics. Been supporting Janine Illian teaching inlabru and teaching disagapp for disaggregation regression. inlabru-courses.github.io/Tanzania_202... Now get the train back to Dar es Salaam

View from train from Dar es Salaam to Dodoma.

130

Stats For Bios @statsforbios.bsky.social · 17/09/2026

Bibbly-bobbly-boo. I am a serious scientist honest.

000

Stats For Bios @statsforbios.bsky.social · 16/09/2026

Reminder that "wiggly" is absolutely a technical statistical term. No discussion needed.

23111

Stats For Bios @statsforbios.bsky.social · 14/09/2026

Not a full analysis example, but I've used clustering to create spatial cross validation folds in datasets where other approaches just acted quite weird.

030

Reposted by Stats For Bios

Stats For Bios @statsforbios.bsky.social · 11/09/2026

Preprint shinyapp by @simonsmart.bsky.social here. User-friendly, reproducible software for high-resolution predictions from aggregate polygon case counts (or other data). Once you're done, click a button to get a Rmd file out. #RStats #DiseaseMapping #malaria www.medrxiv.org/content/10.6...

164

Reposted by Stats For Bios

Francisco Rodriguez-Sanchez @frodsan.bsky.social · 11/09/2026

This paper is 20 years old but keeps being as relevant today as back then :/ doi.org/10.1111/j.13... #statistics #ecopubs

Why do we still use stepwise modelling in ecology and behaviour?

The principal drawbacks of stepwise multiple regression include bias in parameter estimation, inconsistencies among model selection algorithms, an inherent (but often overlooked) problem of multiple hypothesis testing, and an inappropriate focus or reliance on a single best model.

1224

Reposted by Stats For Bios

Richard McElreath 🐈‍⬛ @rmcelreath.bsky.social · 10/09/2026

Modeling non-response is an example I like to use when arguing that "descriptive" statistics also require causal inference. To describe the population, you have to model what causes the sample. Similar to occupancy modeling in ecology in many ways.

110219

Stats For Bios @statsforbios.bsky.social · 11/09/2026

164

Stats For Bios @statsforbios.bsky.social · 10/09/2026

My bar for genAI has been for a while "I'll consider it useful when it can fill in a form for me". Copilot, with a lot of prompting and nonsense, filled 90% of a form for me. That's it, I'm in.

010

Stats For Bios @statsforbios.bsky.social · 09/09/2026

The YouTube android app is just so much worse than the mobile webpage. And I can't delete it.

010

Reposted by Stats For Bios

Aki Vehtari @avehtari.bsky.social · 08/09/2026

I'm looking for doctoral students, postdocs and research fellows to work with on Bayesian Workflow. Flexible starting time

02115

Stats For Bios @statsforbios.bsky.social · 08/09/2026

It will give you a lower performance metric, but won't reduce overfitting. If you use CV to select hyperparameters, *that* will tend to reduce overfitting.

050

Stats For Bios @statsforbios.bsky.social · 08/09/2026

Say I've got a train dataset and a test data set. I fit a linear model to the train set and evaluate within-sample error. Then I fit the linear model to the train set and evaluate with the test set (i.e. cross validation). The 2 models are identical and equally overfit. The metric is different.

192

Stats For Bios @statsforbios.bsky.social · 08/09/2026

I've probably moaned about this before. Cross validation doesn't control overfitting (your model is still just as overfit). It (tries to) estimates predictive performance without being tricked by overfitting. Big difference.

191

Stats For Bios @statsforbios.bsky.social · 06/09/2026

Never seen internal combustion engine as ICE! I was sure you were talking about the USA scumbags.

100

Reposted by Stats For Bios

Visruth Srimath Kandali @visruth.com · 01/09/2026

`stanflow` v0.2.0 is out! `stanflow` is a metapackage a la `tidyverse` to provide a convenient way to install and manage #rstats Stan packages. The package is still in early stages; as such, if you want some features they can probably be added! Stan forum: discourse.mc-stan.org/t/stanflow-v...

discourse.mc-stan.org

`stanflow` v0.2.0 is out

It’s been a while, but stanflow v0.2.0 has just been released by @Visruth. stanflow is a metapackage a la tidyverse to provide a convenient way to install and manage R Stan packages. (See Soft relea...

0166

Stats For Bios @statsforbios.bsky.social · 02/09/2026

Does bioconducter count? Because if so there's been one for a long time

000

Reposted by Stats For Bios

Dave Hemprich-Bennett @davehb.bsky.social · 02/09/2026

It's pretty wild to see how much the air quality has improved in the areas of London I frequented before I left. Looks like NO2 is now ~1/4 of what it used to be at my old uni when I started there. Fantastic stuff

082

Stats For Bios @statsforbios.bsky.social · 01/09/2026

Histogram of number of pages for your viewing pleasure. The 2000 page one is an error, but the other big ones are fairly accurate.

Histogram of number of pages in each book. Quite strong skew. Mode at about 400. A few greater than 1000 and one at around 2000.

010

Stats For Bios @statsforbios.bsky.social · 01/09/2026

Playing around with my Goodreads data. lm(My.Rating ~ Number.of.Pages + Year.Published + Author.Reads + Year.Added) Author.Reads is how many books by that author I've read. Anyone wanna play guess the R^2 or guess sign and sig of variables? #RStats

120

Reposted by Stats For Bios

Francisco Rodriguez-Sanchez @frodsan.bsky.social · 01/09/2026

Great collection of Shiny apps (and R package) for #teaching #ecology. Apps come in different languages (including Spanish) ecoevoapps.gitlab.io

Partial list of Shiny apps designed for teaching ecology, hosted at https://ecoevoapps.gitlab.io/

02418

Reposted by Stats For Bios

medRxivpreprint @medrxivpreprint.bsky.social · 18/06/2026

Accounting for Human Movement to Improve Exposure-Health Models www.medrxiv.org/content/10.64898/20…

001

Stats For Bios @statsforbios.bsky.social · 28/08/2026

Who looks for anything in file properties other than like file size, file type, default app and permissions?

000

Stats For Bios @statsforbios.bsky.social · 28/08/2026

Wanted to know how many unhidden slides I had in my ppt deck. Found a useful solution (right click the file, properties. has slides and hidden slides). The solution is given in such a smug way though! Duh, just look in the file properties. learn.microsoft.com/en-us/answer...

learn.microsoft.com

how to count unhidden slide in power point 365 - Microsoft Q&A

how to count unhidden slide in power point 365?? still use VBA ?

100

Reposted by Stats For Bios

Stats For Bios @statsforbios.bsky.social · 25/08/2026

Reread my blog post on the biggest problem in statistics. On re-read, I still don't think it is terrible (other than the typos and things). So I'll repost. timcdlucas.github.io/the-biggest-...

timcdlucas.github.io

The Biggest Problem In Statistics

03916

Stats For Bios @statsforbios.bsky.social · 28/08/2026

Me and Rohan Arambepola wrote this ages ago. We didn't get it published and then both ran out of time energy to do anything with it. But I think it's still useful. So it can live here now.

timcdlucas.github.io

Spatial_models_with_a_causal_lens

000

Stats For Bios @statsforbios.bsky.social · 28/08/2026

Using a causal lens to help understand spatial models. No maths. Particularly relevant to spatial epidemiologists and species distribution modellers. But others as well. timcdlucas.github.io/spatial_mode... #SDM #malaria #GIS #episky #CausalInference

timcdlucas.github.io

Spatial_models_with_a_causal_lens

252

Stats For Bios @statsforbios.bsky.social · 26/08/2026

"Laughter has dug more useful tunnels all by itself than all the tears on earth." Cortázar - Hopscotch Not sure if I agree but I like it nonetheless.

000

Reposted by Stats For Bios

Stats For Bios @statsforbios.bsky.social · 07/08/2026

Estimating the effect of short term (hourly) air pollution is difficult but that's what we tried to do here. We used data that, unusually, has admission time. We found significant associations between NO2 and asthma admission at 3-6 hour lags but not other lags. bmjopen.bmj.com/content/16/8...

The relative risk (RR) of emergency hospital admissions associated with mean nitrogen dioxide at the specific lags from 0 to 24 hours. Note: the red line shows the average RR of emergency hospital admissions for acute asthma, and the error bars are the 95% CIs.

Lags between 0 and 2 hours are insignificant, with large uncertainties at 0 hours. Lags between 3 and 6 hours are significantly positive, with CIs around 1,005 and 1.04. After 6 hours, all CIs cover zero again.

101

Stats For Bios @statsforbios.bsky.social · 25/08/2026

We mention a few in our paper on unit tests in epidemiology. "bugs caused the Mars Climate Orbiter and the Mariner 1 spacecraft to become lost or destroyed" fits your brief I think. www.sciencedirect.com/science/arti...

sciencedirect.com

Responsible modelling: Unit testing for infectious disease epidemiology

Infectious disease epidemiology is increasingly reliant on large-scale computation and inference. Models have guided health policy for epidemics inclu…

020

Stats For Bios @statsforbios.bsky.social · 25/08/2026

Reread my blog post on the biggest problem in statistics. On re-read, I still don't think it is terrible (other than the typos and things). So I'll repost. timcdlucas.github.io/the-biggest-...

timcdlucas.github.io

The Biggest Problem In Statistics

03916

Reposted by Stats For Bios

Shikhara Bhat @shikharabhat.bsky.social · 20/08/2026

Some birds from Sabah to brighten your timeline

0131

Stats For Bios @statsforbios.bsky.social · 13/08/2026

Album of the week. A black metal (blackgaze really) album about Bill the Pony. I mean look at that cover.

A cute cartoony illustration of Bill the Pony.

050