Lightnews — Scholar-powered news

Laurent Bergé @lrberge.bsky.social · 8d

I feel seen!

Reposted by Laurent Bergé

Libby Heeren @libbyheeren.bsky.social · 17d

Tell me something you do when you code that other people would tell you that you shouldn't do.

Tell me the rules you break!

I'll go first: I work in untitled files in the wrong project directories all the time. Like, all the time. Yes, I do tend to lose things 😂 #databs #rstats #python

78 26 150

Laurent Bergé @lrberge.bsky.social · 16d

I use =

5 1 25

Reposted by Laurent Bergé

Grant McDermott @gmcd.bsky.social · 18d

`tinyplot` 0.5.0 is now available from CRAN & R-universe. Alongside bug fixes & internal improvements, this release includes some cool new features like bubble chart + dual legend support, element dodging, & ephemeral themes.

Detailed changelog: grantmcdermott.com/tinyplot/NEW...

#rstats #dataviz

tinyplot bubble chart example with dual legend, using penguins data. Taken from https://grantmcdermott.com/tinyplot/vignettes/gallery.html

1 14 58

Reposted by Laurent Bergé

Susan Vanderplas @srvanderplas.bsky.social · 25d

Sarcasm, but actually something brought up by one of the deans, that having statisticians around is *less* important in the age of AI because AI can just do experimental design and data analysis for researchers. No joke. #sendhelp

7 9 42

Reposted by Laurent Bergé

Emily Riederer @emilyriederer.bsky.social · 26d

Really insightful post from Julie Tibshirani (spotted in LinkedIn, can't find on Bsky) reflecting on #rstats 's unique governance structure and what can be learned for other languages

jtibs.substack.com/p/if-all-the...

If all the world were a monorepo

The R ecosystem and the case for extreme empathy in software maintenance

jtibs.substack.com

7 50 130

Laurent Bergé @lrberge.bsky.social · 28d

+100

I don't understand this need to overclaim.

I they want to create a new pkg doing stg already existing: sure no problem, since variety is good.

But falsely claiming that the original pkg is bad.... really incorrect.
And it's typical alas...

2

Reposted by Laurent Bergé

Grant McDermott @gmcd.bsky.social · 29d

Important note for Mac users: make sure your installation properly exploits `fixest` multi-threading capabilities:

1. Follow @coatless.bsky.social's simple shell script and OpenMP instructions here: github.com/coatless-she...

2. Install from source: `install.packages("fixest", type = "source")`

1 2 4

Laurent Bergé @lrberge.bsky.social · Sep 10

Among the new features:
- new builtin VCOV: HC2 aand HC3 (thanks to Kyle Butts!)
- new function sparse_model_matrix
- improved internal algorithms
- coefplot/iplot become similar to etable
- batched estimations are easier to make

& many many more, check out the news file!
github.com/lrberge/fixe...

1 5

Laurent Bergé @lrberge.bsky.social · Sep 10

#rstats #econsky

fixest v0.13.0 is finally out!

It's still about making OLS and GLM estimations easy.

Some major changes:
- *default* VCOV becomes iid always!
- singletons are removed by default!

See all the changes here:
github.com/lrberge/fixe...

fixest: Fast Fixed-Effects Estimations

Fast and user-friendly estimation of econometric models with multiple fixed-effects. Includes ordinary least squares (OLS), generalized linear models (GLM) and the negative binomial. The core of the p...

cran.r-project.org

2 9 40

Laurent Bergé @lrberge.bsky.social · Aug 24

The EEA conference starts tomorrow in Bordeaux!
I'll be around, so if you want to meet, just reach out!

#rstats #econsky

www.eeassoc.org/events/eea-b...

This link will take you to a page that’s not on LinkedIn

lnkd.in

1 2

Laurent Bergé @lrberge.bsky.social · Jul 25

Yep, this is not native.
I had to develop my own VSCode extension to make it work.

Reposted by Laurent Bergé

Grant McDermott @gmcd.bsky.social · Jul 18

#rstats #econsky PSA: The next release of `fixest` will include some important changes (plus cool new features).

E.g. Fixed-effects regs will now default to 'iid' SEs rather than clustered. github.com/lrberge/fixe...

You can install and test drive the dev version from R-universe; see the README.

From the fixest NEWS file: the new default VCOV is iid for all estimations. To change the default to the way it was, place setFixest_vcov(all = "cluster", no_FE = "iid") in your .Rprofile.

2 12 33

Laurent Bergé @lrberge.bsky.social · Jul 6

For the first use, there is predict.mira that does not seem to exist, so it can be created no problem.

For the second usage, it seems that MI is done to the predict method of a non-MI estimation.
=> I don't see the issue of creating a new function here bc the use case is very specific

1 1

Laurent Bergé @lrberge.bsky.social · Jul 4

In general masking any base R (including stats, utils, etc) function is a bad idea bc other packages rely on them and expect them to work in a very specific way.

I don't think what you want is possible because of the way S3 works.
Maybe change the design? It might be the easiest way.

1 3

Laurent Bergé @lrberge.bsky.social · Jul 2

Announcing sircon: a new R console! (for Windows)

It's in beta and actively developed (new patched version just released).

I've been using it for a while now and I think it's rather bug free. But if you encounter some: let me know!

github.com/lrberge/sircon

1 1 3

Laurent Bergé @lrberge.bsky.social · Jul 2

It's in beta though but I'm actively developing it. Comments welcome!

BTW to install:
remotes::install_github("https://github.com/lrberge/rmake")

1

Laurent Bergé @lrberge.bsky.social · Jul 2

Hey, I'm new developing a package to make reproductibility easy.

It's rmake!

Contrary to {targets}, you don't need to set up anything, it is all automatic!

So you can have modular code while writing in a single script!!!

github.com/lrberge/rmake

2 1 3

Laurent Bergé @lrberge.bsky.social · Jun 24

OK, thanks for your answer.

You even comment your gitignore :-), nice! (I never do that, but I should!)

Laurent Bergé @lrberge.bsky.social · Jun 23

This fool's actions may, in the end, just favor the renewable energy sector (due to the change in relative prices).

I love the irony!

2

Laurent Bergé @lrberge.bsky.social · Jun 23

Sorry for all those questions, thanks for sharing!

1 2

Laurent Bergé @lrberge.bsky.social · Jun 23

Did you like the quarto workflow (vàv other ways to create such websites)? Did you cut dev time by using it?

Very nice looking btw, well done!

1 3

Laurent Bergé @lrberge.bsky.social · Jun 20

Haha, thanks Grant! But TBH I don't know how soon that will be!

2

Laurent Bergé @lrberge.bsky.social · Jun 20

There are a few productivity bonuses at the end of the presentation (which was fully written in typst btw), in particular:

- a new way to handle reproductibility in R
- a new R console (!)

Although both projects are in active dev, they work:

github.com/lrberge/rmake

github.com/lrberge/sircon

GitHub - lrberge/rmake

Contribute to lrberge/rmake development by creating an account on GitHub.

github.com

1 2 10

Laurent Bergé @lrberge.bsky.social · Jun 20

Just gave a presentation a the REGIS summer school we organised in Bordeaux.

Algorithms for empirical research

Slides and material here: github.com/lrberge/2025...

The conclusion is that, even in the age of AI, using your brain is still relevant!

#RStats #EconSky

2 5 12