banner
clibassi.bsky.social
CJ Libassi
@clibassi.bsky.social
880 followers 530 following 80 posts
phd student in econ and ed at EPSAatTC. formerly: SMPAGWU, College Board, CAPhighered, edpolicyford, ComunidadMadrid, pgcps.
Posts Media Videos Starter Packs
Reposted by CJ Libassi
crampell.bsky.social
Catherine Rampell @crampell.bsky.social · 8d
Whoa, White House withdraws Trump’s controversial nominee to lead BLS after ousting predecessor over jobs data
www.cnn.com/2025/09/30/p...
White House withdraws Trump’s controversial nominee to lead BLS after ousting predecessor over jobs data | CNN Politics
The White House has sent paperwork to the Senate to withdraw the nomination of E.J. Antoni as head of the Bureau of Labor Statistics, three sources told CNN.
www.cnn.com
24 110 550
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · 13d
How about this? drive.google.com/file/d/1QdMA...

Unarchiver was able to get it to extract for me.
drive.google.com
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · 13d
Blast! I may have it on an old external hard drive - I can check later tonight
1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · 13d
Does the file at this link work for you? web.archive.org/web/20210213...
web.archive.org
1
Reposted by CJ Libassi
benzipperer.org
Ben Zipperer @benzipperer.org · Aug 27
one amazing feature here is simply loading the entire dataset would probably freeze your laptop!

instead you can run this regression quickly and not worry about memory problems, thanks to the magic of duckdb and dbreg
gmcd.bsky.social
Grant McDermott @gmcd.bsky.social · Aug 25
To borrow another example, taken from the `dbreg` README: github.com/grantmcdermo...

Here I am running a fixed-effects regression on 180 million(!) row parquet dataset... and it completes **< 2 seconds**... on my laptop 🤯

This is powered by @duckdb.org under the hood.

#rstats #econsky
Running dbreg::dbreg() on a full year of NYC taxi data... and it takes less than 2 seconds. dbreg( tip_amount ~ fare_amount + passenger_count | month + vendor_name, path = "read_parquet('nyc-taxi/**/*.parquet')", ## path to hive-partitioned dataset vcov = "hc1" ) #> [dbreg] Estimating compression ratio... #> [dbreg] Data has 178,544,324 rows and 24 unique FE groups. #> [dbreg] Using strategy: compress #> [dbreg] Executing compress strategy SQL #> #> Compressed OLS estimation, Dep. Var.: tip_amount #> Observations.: 178,544,324 (original) | 70,782 (compressed) #> Standard Errors: Heteroskedasticity-robust #> Estimate Std. Error t value Pr(>|t|) #> fare_amount 0.106744 0.000068 1564.742 < 2.2e-16 *** #> passenger_count -0.029086 0.000106 -273.866 < 2.2e-16 *** #> --- #> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
1 5 24
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 20
Anyone know if anyone is doing any formal benchmarking of LLM performance on Stata tasks? www.statalist.org/forums/forum...
Anyone running benchmarks across LLMs for Stata skill? - Statalist
Curious if anyone has already tried comparing how well different LLMs perform on Stata coding tasks—not just casually, but using formal benchmarks. I’m
www.statalist.org
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 18
Certainly not arguing the status quo (where all sorts of award letter shenanigans can and do occur) is optimal, but just as a matter of calculation, how would you calculate the value of IDR, PSLF, borrower protections such as hardship forbearances, in school deferments, closed school discharge, etc.
1 1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 16
Thanks for all you do for this package!
1
Reposted by CJ Libassi
s3alfisc.bsky.social
Alex @s3alfisc.bsky.social · Aug 16
Vis method for decomposition now merged to main, feedback welcome!
2 1 6
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 12
Also could it handle “negative” contributions (variables that increase the coefficient) in a way that is visually intuitive?
1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 12
This looks great!! Think it would scale well to many covariates?
1 1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Aug 12
Man, when @dieworkwear.bsky.social weighs in on EJ Antoni's outfits, it may shake the earth.
1
Reposted by CJ Libassi
erikamcentarfer.bsky.social
Erika McEntarfer @erikamcentarfer.bsky.social · Aug 2
It has been the honor of my life to serve as Commissioner of BLS alongside the many dedicated civil servants tasked with measuring a vast and dynamic economy. It is vital and important work and I thank them for their service to this nation.
1.2K 4.4K 22K
Reposted by CJ Libassi
copafs.bsky.social
COPAFS @copafs.bsky.social · Aug 2
AEA Statement on Dismissal of BLS Comm.

"The independence of the federal statistical agencies is essential to the proper functioning of a modern economy. Accurate, timely, and impartial statistics are the foundation upon which households, businesses, and policymakers make critical decisions."
3 110 200
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jul 31
Thanks for maintaining it! Was very grateful to see that it was part of pyfixest.
1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jul 31
I think these look great! Very logical way to put things together. The challenge in my mind is how to handle many vars? One thing I have toyed with for this is trying to plot the top N vars decomposition results. Something like this toy example I just had Claude code whip up on simulated data.
2 2
Reposted by CJ Libassi
riacton.bsky.social
Riley Acton @riacton.bsky.social · Jul 29
Excited to announce the call for papers for the inaugural MidSouth Education Policy Workshop, October 16-17, in Lexington, KY!

Send us your abstracts on all things econ of ed & ed policy by 8/27. Grad students & early career folks especially welcome!

Info & link to submit here: bit.ly/44TdiGf
Lexington, KY
1 25 28
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jul 8
ICYMI last week - take a peek at our new report to better understand the earnings prospects of the professional school programs Congress just (nearly) uniformly & dramatically changed liquidity provision for: pseocoalition.org/wp-content/u...
1
Reposted by CJ Libassi
claremccann.bsky.social
Clare McCann @claremccann.bsky.social · Jul 2
@jdmatsudaira.bsky.social and @clibassi.bsky.social explain what new graduate loan limits will mean, not only for borrowers in medical school but across the graduate education landscape. Check out this thread/follow @peerresearch.bsky.social for more!
Jordan Matsudaira @jdmatsudaira.bsky.social · Jul 2
House/Senate are set to eliminate Grad PLUS, and put in place loan limits that would dramatically restrict graduate borrowing. Media coverage emphasizes impact this may have on very expensive programs like Medicine and Dentistry, but @peerresearch.bsky.social finds impact will be much(!) broader 1/
1 3
Reposted by CJ Libassi
Jordan Matsudaira @jdmatsudaira.bsky.social · Jul 2
House/Senate are set to eliminate Grad PLUS, and put in place loan limits that would dramatically restrict graduate borrowing. Media coverage emphasizes impact this may have on very expensive programs like Medicine and Dentistry, but @peerresearch.bsky.social finds impact will be much(!) broader 1/
3 33 49
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jun 30
Truly thank you. Finding these emojis was the most exciting part of this.
2
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jun 30
Much more in the report, from tracing earnings trajectories across programs to how earnings paths interact with IDR. We hope you will take a read! Many thanks to the folks who volunteered to give comments on drafts, helping improve the report tons. Find it here:: pseocoalition.org/wp-content/u...
pseocoalition.org
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jun 30
On top of tracking debt-to-earnings over a longer period, we look at different parts of the distribution of earnings. Here for example is the interquartile range of earnings for all of the students in the PSEO data, which captures a recurring fact in the report: things are just different in medicine
Chart showing earnings ranges for professional programs 5 and 10 years after graduation. Horizontal bars display interquartile ranges with median lines. Programs from lowest to highest 10-year median earnings: Rehabilitation/Therapeutic Professions, Veterinary Medicine, Optometry, Law/JD, Pharmacy, Dentistry, and Medicine/MD. Medicine shows dramatically higher earnings than other programs, with 10-year median above $300k and upper range extending to over $500k. Most other programs cluster between roughly $100k-$150k for 10-year median earnings
1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jun 30
Maybe the biggest headline is just how much higher earnings for these professional programs is than the (quite large!) debt levels reported by ED. Remember that this is over just the first 10 years following graduation, and we still see ratios of total earnings to total debt near or exceeding 10x.
1
clibassi.bsky.social
CJ Libassi @clibassi.bsky.social · Jun 30
New at @pseocoalition.bsky.social, @julia-turner.bsky.social & I have a new report on grad school debt & earnings over the medium term. For some key professional fields (🩺⚖️🦷💊🐾), we show the varied patterns both within & across areas of study, looking at the first decade of earnings after graduation
Scatter plot comparing median program debt (x-axis, $50k-$300k) vs 10-year cumulative earnings (y-axis, $0.5M-$2M) for professional programs. Medicine MD programs (pink squares) cluster in upper right with high debt ($150k-$250k) and high earnings ($1.4M-$1.8M). Law programs (blue circles) cluster in the bottom left corner with lower debt and lower earnings, though some elite programs show medicine-level earnings and somewhat higher debt than other law schools. Veterinary Medicine (green diamonds), Dentistry (plus signs) and Pharmacy (X marks) are distributed across middle ranges of earnings, but across wide ranges of the distribution of debt, with almost all pharmacy programs having lower debt than almost all dentistry programs. Physical therapy and veterinary programs have law-like earnings (between half a million and a million over 10 years), and while PT has law-like debt as well, veterinary debt is generally much higher and closer to medical school.
3 12 22