Laurent Bergé | Lightnews

Light up
your news

About Privacy Terms Help

Laurent Bergé

Laurent Bergé

@lrberge.bsky.social

500 followers 540 following 130 posts

Ass prof in econ at the University of Bordeaux

Too much code.

https://sites.google.com/site/laurentrberge/

Business 50%

Physics 26%

more

Posts Replies Media Videos

Pinned

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

Laurent Bergé @lrberge.bsky.social · 10d

Félicitations!!! Je ne connaissais pas : c'est compliqué de postuler (je suis moi même contributeur) ?

Laurent Bergé @lrberge.bsky.social · 11d

Thanks for the resource: it looks very nice!

Laurent Bergé @lrberge.bsky.social · 11d

At the beginning of my teacher's career I taught mostly for the best students in the class.

As I get older, I can see I'm lowering my standards and teach more and more for the weaker students.

I also grade more generously as time passes.

=> is this a common pattern? Or is it just me? I'm curious.

Laurent Bergé @lrberge.bsky.social · 11d

Thanks! I knew the other but not this one! Good to know!

Laurent Bergé @lrberge.bsky.social · 11d

Thanks Alex, that's a great ref indeed!

Laurent Bergé @lrberge.bsky.social · 11d

Thanks for sharing!

Laurent Bergé @lrberge.bsky.social · 11d

The website is super cool!

Laurent Bergé @lrberge.bsky.social · 11d

Thanks a lot Achim: that's exactly the target's level!

Just went through the intro: tis great!

Laurent Bergé @lrberge.bsky.social · 11d

Thanks! But starting with the shell, they'll be off right away :-D
I wish I has sciencepo students!!!

Laurent Bergé @lrberge.bsky.social · 11d

Thanks for the references: I'll have a look at the mit gnl prog, it might be the one!

I'll also link them to your course.

Laurent Bergé @lrberge.bsky.social · 11d

Thanks for the ref. Very nice course website btw!

Laurent Bergé @lrberge.bsky.social · 11d

Some context: I do teach an intro to R programming class,
and it goes smoothly for the students with some prior programming knowledge.

However I have some (admittedly weak, but with goodwill) students with no experience in programming and I feel it's an uphill battle to teach them anything.

Laurent Bergé @lrberge.bsky.social · 16d

Any good ressource for LEARNING PROGRAMMING for students who know NOTHING about programming (really nothing)?

It would be great in R/python (final obj. is data analysis).

Any help appreciated! Thanks!

@andrew.heiss.phd @gmcd.bsky.social @kylefbutts.bsky.social @vincentab.bsky.social

Reposted by Laurent Bergé

pikaole @pikaole.bsky.social · 17d

🦀🐚

Two-panel cartoon
1. A small hermit crab praises the shell of a Purple-ringed Topsnail.
2. A year later, a hermit crab, having grown larger and inherited the shell, is grateful.

The atmosphere somehow feels like a law of nature.

Laurent Bergé @lrberge.bsky.social · 19d

lol, we're big bluey fans at home!

Reposted by Paul Goldsmith-Pinkham, Laurent Bergé

Grant McDermott @gmcd.bsky.social · 29d

Happy `tinyplot` v0.6.0 (codename "Thanksgiving") release day to all those that celebrate. Some new features, but mostly bug fixes and internal improvements.
grantmcdermott.com/tinyplot/NEW...

#rstats

Example of dodged pointrange (tiny)plot, with aligned ribbons layered on top.

Laurent Bergé @lrberge.bsky.social · Oct 2

I feel seen!

Laurent Bergé @lrberge.bsky.social · Sep 24

I use =

Reposted by Laurent Bergé

Libby Heeren @libbyheeren.bsky.social · Sep 24

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

Reposted by Paul Goldsmith-Pinkham, Barrett, Laurent Bergé

Grant McDermott @gmcd.bsky.social · Sep 23

`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

Reposted by Laurent Bergé

Susan Vanderplas @srvanderplas.bsky.social · Sep 15

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

Reposted by Barrett, Laurent Bergé, Julen Astigarraga

Emily Riederer @emilyriederer.bsky.social · Sep 14

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

Laurent Bergé @lrberge.bsky.social · Sep 12

+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...

Reposted by Laurent Bergé

Grant McDermott @gmcd.bsky.social · Sep 11

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")`

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...

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

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 Ingo Rohlfing, 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.

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

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.