Lightnews — Scholar-powered news

Zhentao Shi @zhentaoshi.bsky.social · Jul 13

It is a good time for reading on the days of sleep disorder (due to jet lag).

Zhentao Shi @zhentaoshi.bsky.social · Jul 13

This is insane. So sad about the loss.

Reposted by Zhentao Shi

Martin Weidner @weidnerecon.bsky.social · Jun 5

Shameless plug: Still time to join the World Congress of the Econometric Society (Aug 18–22, Seoul)! Stellar invited speakers 👉 www.eswc2025.org/speakers/03....
Even the keynotes look pretty sharp 👉 www.eswc2025.org/speakers/01....
Register by June 30!

ESWC 2025

www.eswc2025.org

1 3

Zhentao Shi @zhentaoshi.bsky.social · Apr 3

The world is not what I knew before.

Zhentao Shi @zhentaoshi.bsky.social · Jan 16

The trend for business and economics education is not going well. Changes are needed.

Even Harvard M.B.A.s Are Struggling to Land Jobs

The latest crop of elite business-school graduates is taking months to find new jobs.

www.wsj.com

Zhentao Shi @zhentaoshi.bsky.social · Jan 16

The element-by-element is indeed the default option of `*` in many languages, such as Python and R.

MathMatize Memes @mathmatize.bsky.social · Jan 16

Cursed matrix multiplication

#MathSky

$2 characters carrying another character in a stretcher. The characters are saying "What happened to him?" and "He found out (3&6\\0&1)(5&4\\0&2)=(3\cdot 5&6\cdot 4 \\ 0\cdot 0&1\cdot2)". The character in the stretcher is captioned "Mathematician".$

Zhentao Shi @zhentaoshi.bsky.social · Jan 13

If a discipline labels itself as "science", it is pseudoscience.

We don't call physics "physical science", nor chemistry "chemical science".

Zhentao Shi @zhentaoshi.bsky.social · Jan 8

To start the new semester's teaching tomorrow. Still recovering from jet lag of 8 hours to the west.

Zhentao Shi @zhentaoshi.bsky.social · Jan 8

Greenland, Canada, and Panama. Interesting.

Zhentao Shi @zhentaoshi.bsky.social · Jan 7

Returned HK from SFO for ASSA 2025. Met with lots of old friends.

Reposted by Zhentao Shi

Pietro Biroli @pietrobiroli.bsky.social · Dec 14

I'm not an R person, but I totally agree with this point: we're doing students a disservice by teaching them stata.

Even worse if you're teaching at a public university: using public money to create consumers for a private company.

Josh Merfeld @merfeld.bsky.social · Dec 14

We should care about students’ career outcomes. Teaching them R will be MUCH more helpful than Stata. Not only is R more highly valued, it also makes learning things like Python much easier.

10 25 260

Zhentao Shi @zhentaoshi.bsky.social · Dec 6

In the era of capable LLM, writing econometric textbooks will become obsolete. The basic facts about models, estimation methods can be largely covered by LLM, which are getting better day by day.
A textbook author may only be helpful in connecting the dots into a systematic body of knowledge.
#llm

Zhentao Shi @zhentaoshi.bsky.social · Dec 6

With the recent exposure, I trust the potential of LLM in econometrics.

#ai #llm

Zhentao Shi @zhentaoshi.bsky.social · Dec 4

It was a Korean drama. Fortunately it ended very shortly after it had unfolded.

Zhentao Shi @zhentaoshi.bsky.social · Dec 4

It helps with the transition from Stata to a proper data science programming languages, for example R or Py.
Stata has no future.
Stop using it.

Paul Goldsmith-Pinkham @paulgp.com · Dec 4

Just a reminder that @gmcd.bsky.social and @kylefbutts.bsky.social have made an incredible public good describing how to use data.table and fixest to encourage moving from Stata to R!

stata2r.github.io

Translating Stata to R

Learning R coming from Stata

stata2r.github.io

1 2

Zhentao Shi @zhentaoshi.bsky.social · Dec 3

Again, it's the junior job market season.

Reposted by Zhentao Shi

John Holbein @johnholbein1.bsky.social · Nov 30

5 41 370

Reposted by Zhentao Shi

Ben Golub @bengolub.bsky.social · Dec 1

delighted to see this prize go to Arun Chandrasekhar, a pioneer in applications and theory of networks in economics and social science more broadly

a fantastically talented and energetic scholar

2 18 140

Reposted by Zhentao Shi

Peter Reinhard Hansen @profphansen.bsky.social · Nov 28

Without capitalism, there would be no Bluesky to blame capitalism for xyz…

But it’s true. We would not have had a climate crisis, had it not been for capitalism lifting billions out of extreme poverty and keeping us alive longer.

1 1

Zhentao Shi @zhentaoshi.bsky.social · Nov 27

VS code puts the best of two into one environment.

Zhentao Shi @zhentaoshi.bsky.social · Nov 27

Jupyter notebook is a nice format for presenting code, but according to my personal experience not the best format for developing code.

Perhaps I am old-styled. I like a traditional, full functional IDE better.

1

Reposted by Zhentao Shi

Paul Goldsmith-Pinkham @paulgp.com · Nov 7

Pleased with this one:

Spatial Unit Roots and Spurious Regression (Mueller and Watson (2024))
paulgp.github.io/2024/11/06/r...

I asked it an additional follow-up question based on my teaching of Kelly (2020) on Spatial Persistence as well, and it gave a wonderful set of answers.

2 7 39

Reposted by Zhentao Shi

Anne Applebaum @anneapplebaum.bsky.social · Nov 27

People are often surprised when I tell them this: both Twitter and Facebook/insta/threads are now specifically engineered to downgrade, hide, suppress real journalism. If this platform is different, that's a huge change

Matt Karolian @mkarolian.bsky.social · Nov 26

Traffic from Bluesky to @bostonglobe.com is already 3x that of Threads, and we are seeing 4.5x the conversions to paying digital subscribers.

460 7.7K 38K

Zhentao Shi @zhentaoshi.bsky.social · Nov 26

At time time, the Internet will be the forum for AI to debate about cyber politics. Journals (if their still are such things) will be the outlets for AI to write and review findings.

There will be no place for human, except manual labors that AI are still not good at (yet).

1

Zhentao Shi @zhentaoshi.bsky.social · Nov 26

In the future if most of the text is written by AI, human researchers will be inundated.

What's the point of reading papers, if AI generates papers and peer-review them?

1 1