iiinova — where your digital self keeps evolving

Thomas Blanchet

@thomasblanchet.bsky.social

Data Scientist/Software Engineer in industry | Recovering academic | Formerly economist at UC Berkeley (Postdoc), Paris School of Economics (PhD), ENSAE ParisTech (MSc)

Posts Replies Media

Thomas Blanchet @thomasblanchet.bsky.social · 11/12/2024

But what if your theory that more supply reduces prices is wrong and we just end up building a bunch of modern, environmentally efficient dwellings for people in places with a lot of amenities and economic opportunities for nothing?

1101

Thomas Blanchet @thomasblanchet.bsky.social · 26/11/2024

I gave up on academia after seeing a few too many of: "I made up that number, but will only make that clear in some obscure appendix, giving me plausible deniability against accusations of fraud, while betting on enough readers to be too lazy to question it, so that the figure becomes widely cited."

020

Thomas Blanchet @thomasblanchet.bsky.social · 23/11/2024

People like RDDs because it's a pre-approved method that's good at turning noise into flashy results. Academia selects against people who think about whether such tools make sense in context because doing so will lower your paper output.

000

Thomas Blanchet @thomasblanchet.bsky.social · 23/11/2024

But the RDD graph clearly shows that a basic regression would show a lower effect here. So, what are we arguing? That places more likely to elect women mayors have worse domestic violence?

100

Thomas Blanchet @thomasblanchet.bsky.social · 23/11/2024

After all, why can't you simply regress domestic violence on a woman mayor dummy? Because places with women mayors probably have better gender norms all around, which would lower domestic violence. So that regression would overstate the true effect.

100

Thomas Blanchet @thomasblanchet.bsky.social · 23/11/2024

The claim that having a woman mayor cuts domestic homicides *by half* obviously absurd. But this paper is also a great example of people forgetting why RDDs exist in the first place: www.sciencedirect.com/science/arti...

sciencedirect.com

Fighting violence against women: The role of female political representation

We investigate the effect of women’s political leadership in local government on domestic violence against women. Using a regression discontinuity des…

100

Thomas Blanchet @thomasblanchet.bsky.social · 23/11/2024

Economists will literally believe there is a discontinuity here instead of going to therapy.

You're not missing much, it's a regression discontinuity graph with no actual discontinuity.

150