econmaett
@econmaett.github.io
❤️📊 | 🗣️DE|EN|FR | #rstats | #econsky
Reposted by econmaett
Come see me talk about Positron and Raycast and all sorts of fun things to make your data analysis life better!
TOMORROW! (Tuesday)
@andrew.heiss.phd is gonna be on the DS Lab showing us how he makes his life easier: #Positron extensions he loves, OS tools he uses, and #duckdb + Connections pane 👀
Bonus, we'll def share the "moving the activity bar" thing live, too 👏🏻
Register: pos.it/dslab
#rstats #databs
@andrew.heiss.phd is gonna be on the DS Lab showing us how he makes his life easier: #Positron extensions he loves, OS tools he uses, and #duckdb + Connections pane 👀
Bonus, we'll def share the "moving the activity bar" thing live, too 👏🏻
Register: pos.it/dslab
#rstats #databs
January 12, 2026 at 11:06 PM
Come see me talk about Positron and Raycast and all sorts of fun things to make your data analysis life better!
Reposted by econmaett
Hesitating very much to put this on the internet, but: any suggestions for guides for exploring Claude code for applied micro analysis (these days I mostly work on RCTs); primarily in R; not a current Github user (I'm open to changing); AND step-by-step from very beginning
January 12, 2026 at 10:53 PM
Reposted by econmaett
I mean, he’s not wrong
January 12, 2026 at 10:26 PM
I mean, he’s not wrong
Reposted by econmaett
I'm impressed I was so consistent with myself! I for sure thought I'd be inconsistent somewhere
Also I'm keen to see an analysis of the data - I did a bunch of reading on this area to inform some recommendations around probability number -> wording
This area is so interesting!
Also I'm keen to see an analysis of the data - I did a bunch of reading on this area to inform some recommendations around probability number -> wording
This area is so interesting!
January 12, 2026 at 9:35 PM
I'm impressed I was so consistent with myself! I for sure thought I'd be inconsistent somewhere
Also I'm keen to see an analysis of the data - I did a bunch of reading on this area to inform some recommendations around probability number -> wording
This area is so interesting!
Also I'm keen to see an analysis of the data - I did a bunch of reading on this area to inform some recommendations around probability number -> wording
This area is so interesting!
Reposted by econmaett
i'm too lazy to update a powerpoint every month.
so i guess we're gonna learn how to work with xml in R so we can automate this sucker.
what lazy task is motivating you to learn new things today?
so i guess we're gonna learn how to work with xml in R so we can automate this sucker.
what lazy task is motivating you to learn new things today?
January 12, 2026 at 8:09 PM
i'm too lazy to update a powerpoint every month.
so i guess we're gonna learn how to work with xml in R so we can automate this sucker.
what lazy task is motivating you to learn new things today?
so i guess we're gonna learn how to work with xml in R so we can automate this sucker.
what lazy task is motivating you to learn new things today?
Reposted by econmaett
#rstats tip: you probably don't realize how *slow* pmax() is.
Which is fine if speed doesn't matter, but I just profiled some maths code using Airy functions and numerical integration and stuff, and
n<-pmax(n,1)
was taking like a quarter of the time. You can use
n[n<1] <- 1
if that's all you need
Which is fine if speed doesn't matter, but I just profiled some maths code using Airy functions and numerical integration and stuff, and
n<-pmax(n,1)
was taking like a quarter of the time. You can use
n[n<1] <- 1
if that's all you need
January 12, 2026 at 10:30 PM
#rstats tip: you probably don't realize how *slow* pmax() is.
Which is fine if speed doesn't matter, but I just profiled some maths code using Airy functions and numerical integration and stuff, and
n<-pmax(n,1)
was taking like a quarter of the time. You can use
n[n<1] <- 1
if that's all you need
Which is fine if speed doesn't matter, but I just profiled some maths code using Airy functions and numerical integration and stuff, and
n<-pmax(n,1)
was taking like a quarter of the time. You can use
n[n<1] <- 1
if that's all you need
Reposted by econmaett
I see a lot of econ/socsci Bluesky asking for AI starter advice. IMO the highest impact, low cost tweak to your workflow is training an AI to follow *your* coding/writing prefs & habits.
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
January 12, 2026 at 9:14 PM
I see a lot of econ/socsci Bluesky asking for AI starter advice. IMO the highest impact, low cost tweak to your workflow is training an AI to follow *your* coding/writing prefs & habits.
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
Reposted by econmaett
This is a pretty good idea
I see a lot of econ/socsci Bluesky asking for AI starter advice. IMO the highest impact, low cost tweak to your workflow is training an AI to follow *your* coding/writing prefs & habits.
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
Literally, tell Claude: "Look at <XYZ dirs>. ID common themes, styles & conventions. Write them to a .md file."
January 12, 2026 at 9:18 PM
This is a pretty good idea
Hesitating very much to put this on the internet, but: any suggestions for guides for exploring Claude code for applied micro analysis (these days I mostly work on RCTs); primarily in R; not a current Github user (I'm open to changing); AND step-by-step from very beginning
January 12, 2026 at 10:53 PM
Reposted by econmaett
Hesitating very much to put this on the internet, but: any suggestions for guides for exploring Claude code for applied micro analysis (these days I mostly work on RCTs); primarily in R; not a current Github user (I'm open to changing); AND step-by-step from very beginning
January 12, 2026 at 9:04 PM
Hesitating very much to put this on the internet, but: any suggestions for guides for exploring Claude code for applied micro analysis (these days I mostly work on RCTs); primarily in R; not a current Github user (I'm open to changing); AND step-by-step from very beginning
Reposted by econmaett
Hesitating very much to put this on the internet, but: any suggestions for guides for exploring Claude code for applied micro analysis (these days I mostly work on RCTs); primarily in R; not a current Github user (I'm open to changing); AND step-by-step from very beginning
January 12, 2026 at 10:29 PM
Reposted by econmaett
Thank you! This is nifty!
"hardcore" R package dev skill · Issue #1 · jeremy-allen/claude-skills
Thanks @sj-io for sharing this with me! I think the R package development skill is a bit too "high-level" for my tastes. I'd love to see a more "hardcore" or barebones version. For me this would be...
github.com
January 12, 2026 at 8:32 PM
Thank you! This is nifty!
Reposted by econmaett
I also experienced this and made a CLAUDE.md file last year to help with some of the basics, and @jeremy-data.bsky.social just turned it into Claude Skills! They're in his repo for now but plans to add them to Posit's public skills repo! github.com/jeremy-allen...
GitHub - jeremy-allen/claude-skills
Contribute to jeremy-allen/claude-skills development by creating an account on GitHub.
github.com
January 12, 2026 at 8:13 PM
I also experienced this and made a CLAUDE.md file last year to help with some of the basics, and @jeremy-data.bsky.social just turned it into Claude Skills! They're in his repo for now but plans to add them to Posit's public skills repo! github.com/jeremy-allen...
Reposted by econmaett
New on CRAN: grangersearch (0.1.0). View at https://CRAN.R-project.org/package=grangersearch
grangersearch: Granger Causality Testing for Time Series
Performs Granger causality tests on pairs of time series to determine causal relationships. Uses Vector Autoregressive (VAR) models to test whether one time series helps predict another beyond what the series' own past values provide. Returns structured results including p-values, test statistics, and causality conclusions for both directions.
CRAN.R-project.org
January 12, 2026 at 9:23 PM
New on CRAN: grangersearch (0.1.0). View at https://CRAN.R-project.org/package=grangersearch
Reposted by econmaett
I’m happy to share that the print version of Longitudinal Data Analysis Using R is now available on Amazon — designed for teaching, applied research, and self-study.
👉 www.amazon.com/Longitudinal...
#rstats #HigherEd #Statistics #LongitudinalData #SocialScience #Statistics
👉 www.amazon.com/Longitudinal...
#rstats #HigherEd #Statistics #LongitudinalData #SocialScience #Statistics
Longitudinal Data Analysis Using R
Amazon.com: Longitudinal Data Analysis Using R: 9798261825210: Cernat, Alexandru: Books
www.amazon.com
January 12, 2026 at 3:28 PM
I’m happy to share that the print version of Longitudinal Data Analysis Using R is now available on Amazon — designed for teaching, applied research, and self-study.
👉 www.amazon.com/Longitudinal...
#rstats #HigherEd #Statistics #LongitudinalData #SocialScience #Statistics
👉 www.amazon.com/Longitudinal...
#rstats #HigherEd #Statistics #LongitudinalData #SocialScience #Statistics
Reposted by econmaett
i have the opposite problem where google thinks I need to hear from this dolt 🤦🏻♂️🤣
January 12, 2026 at 6:02 PM
i have the opposite problem where google thinks I need to hear from this dolt 🤦🏻♂️🤣
Reposted by econmaett
This week on What's New in R:
✅ Data viz. on domestic terrorism trends by Stephen Ponce
✅ Ted Laderas introduces Databot, Positron’s AI-powered analysis tool
✅ {quanteda} package developed by Kenneth Benoit for analyzing qualitative text data
Read the issue: buff.ly/gUPhm9s
#rstats
✅ Data viz. on domestic terrorism trends by Stephen Ponce
✅ Ted Laderas introduces Databot, Positron’s AI-powered analysis tool
✅ {quanteda} package developed by Kenneth Benoit for analyzing qualitative text data
Read the issue: buff.ly/gUPhm9s
#rstats
January 12, 2026 at 4:03 PM
This week on What's New in R:
✅ Data viz. on domestic terrorism trends by Stephen Ponce
✅ Ted Laderas introduces Databot, Positron’s AI-powered analysis tool
✅ {quanteda} package developed by Kenneth Benoit for analyzing qualitative text data
Read the issue: buff.ly/gUPhm9s
#rstats
✅ Data viz. on domestic terrorism trends by Stephen Ponce
✅ Ted Laderas introduces Databot, Positron’s AI-powered analysis tool
✅ {quanteda} package developed by Kenneth Benoit for analyzing qualitative text data
Read the issue: buff.ly/gUPhm9s
#rstats
Reposted by econmaett
🎉 shinytest2 v0.5.0 is now on CRAN!
✨ AppDriver now auto-loads local packages - no more installing required!
🔧 Pass functions directly to AppDriver$new(app_dir=). Ex: Golem's run_app()
📦 New local_app_support() & with_app_support() helpers
rstudio.github.io/shinytest2/
#rstats #rshiny
✨ AppDriver now auto-loads local packages - no more installing required!
🔧 Pass functions directly to AppDriver$new(app_dir=). Ex: Golem's run_app()
📦 New local_app_support() & with_app_support() helpers
rstudio.github.io/shinytest2/
#rstats #rshiny
January 12, 2026 at 4:11 PM
🎉 shinytest2 v0.5.0 is now on CRAN!
✨ AppDriver now auto-loads local packages - no more installing required!
🔧 Pass functions directly to AppDriver$new(app_dir=). Ex: Golem's run_app()
📦 New local_app_support() & with_app_support() helpers
rstudio.github.io/shinytest2/
#rstats #rshiny
✨ AppDriver now auto-loads local packages - no more installing required!
🔧 Pass functions directly to AppDriver$new(app_dir=). Ex: Golem's run_app()
📦 New local_app_support() & with_app_support() helpers
rstudio.github.io/shinytest2/
#rstats #rshiny
Reposted by econmaett
📚Coming soon: Chapters from our open-access book "Hello Data Science," based on our intro courses across three institutions!
📦 Today, we’re releasing the early version of the #rstats package with datasets. Find it on GitHub (on its way to CRAN): github.com/hellodata-sc...
📦 Today, we’re releasing the early version of the #rstats package with datasets. Find it on GitHub (on its way to CRAN): github.com/hellodata-sc...
January 12, 2026 at 4:16 PM
📚Coming soon: Chapters from our open-access book "Hello Data Science," based on our intro courses across three institutions!
📦 Today, we’re releasing the early version of the #rstats package with datasets. Find it on GitHub (on its way to CRAN): github.com/hellodata-sc...
📦 Today, we’re releasing the early version of the #rstats package with datasets. Find it on GitHub (on its way to CRAN): github.com/hellodata-sc...
Reposted by econmaett
A nice #RStats irony I noticed today:
📦{grateful} auto-cites all of the packages you used… except itself, unless you set omit = NULL.
Even tools for acknowledgment need acknowledgment. Glad I spotted it and fixed it in time.
pakillo.github.io/grateful/ind...
📦{grateful} auto-cites all of the packages you used… except itself, unless you set omit = NULL.
Even tools for acknowledgment need acknowledgment. Glad I spotted it and fixed it in time.
pakillo.github.io/grateful/ind...
Facilitate Citation of R Packages
Facilitates the citation of R packages used in analysis projects. Scans project for packages used, gets their citations, and produces a document with citations in the preferred bibliography format, re...
pakillo.github.io
January 12, 2026 at 4:54 PM
A nice #RStats irony I noticed today:
📦{grateful} auto-cites all of the packages you used… except itself, unless you set omit = NULL.
Even tools for acknowledgment need acknowledgment. Glad I spotted it and fixed it in time.
pakillo.github.io/grateful/ind...
📦{grateful} auto-cites all of the packages you used… except itself, unless you set omit = NULL.
Even tools for acknowledgment need acknowledgment. Glad I spotted it and fixed it in time.
pakillo.github.io/grateful/ind...
Reposted by econmaett
R-Mob at Charles Sturt Univ (AU) uses #rstats for agronomy + environment.
Best part: “Digital Divide in Agriculture”—students learn data/programming by analyzing family farm records + trial plots. r-consortium.org/posts/crops-...
#AgTech #DigitalAgriculture #DataScience #OpenSource
Best part: “Digital Divide in Agriculture”—students learn data/programming by analyzing family farm records + trial plots. r-consortium.org/posts/crops-...
#AgTech #DigitalAgriculture #DataScience #OpenSource
January 12, 2026 at 5:52 PM
R-Mob at Charles Sturt Univ (AU) uses #rstats for agronomy + environment.
Best part: “Digital Divide in Agriculture”—students learn data/programming by analyzing family farm records + trial plots. r-consortium.org/posts/crops-...
#AgTech #DigitalAgriculture #DataScience #OpenSource
Best part: “Digital Divide in Agriculture”—students learn data/programming by analyzing family farm records + trial plots. r-consortium.org/posts/crops-...
#AgTech #DigitalAgriculture #DataScience #OpenSource
Reposted by econmaett
The way Ukrainian defenders treat animals is a testament to their character and humanity. This humanity is the very reason why we must support Ukraine. “The only thing necessary for the triumph of evil is for good men to do nothing.”
January 12, 2026 at 6:57 PM
The way Ukrainian defenders treat animals is a testament to their character and humanity. This humanity is the very reason why we must support Ukraine. “The only thing necessary for the triumph of evil is for good men to do nothing.”
Reposted by econmaett
From food security to emissions, disappearing data is making it harder to understand what’s really happening in the US.
America’s Statistical System Is Breaking Down
Canceled surveys, missing datasets and staffing cuts are leaving the US with growing blind spots — and weakening trust in official numbers.
bloom.bg
January 12, 2026 at 6:00 PM
From food security to emissions, disappearing data is making it harder to understand what’s really happening in the US.