So five different estimators when we use MLE weights collapse to one estimator using IPT weights. IPWRA and normalized AIPW work pretty well with MLE weights but differ from each other and the other estimators. The IPT-based estimator is hard to beat especially when the mean and PS are both wrong.

ssc install teffects2

The syntax is essentially the same as teffects and allows MLE or IPT logit PS estimation. Where appropriate, we allow normalized or unnormalized weights (with MLE for IPW and AIPW). Preference for normalized. Standard errors account for all estimation uncertainty.

Weights for IPW and AIPW are automatically normalized. Holds for ATE and ATT. With MLE-based weights, the three estimators are all different, and the IPW and AIPW weights are not automatically normalized. We have an accompanying Stata command, teffects2.

Thanks Paul. It’s nice to have a paper that’s both elegant (if I may say so) and practically useful. The conclusion is that using a particular covariate balancing PS estimator — inverse probability tilting — renders IPW, AIPW, and IPWRA all numerically identical with a linear conditional mean.

This puzzled me and seems like a kind of appropriation. Mundlak was squarely in the frequentist/FE camp.

Oh I think you know I’ve always been a barbarian.

We finished days 2-3 of our #summerschool in #healtheconomics

Packed with insights from
@jmwooldridge.bsky.social on recent DiD methods
StefanieSchurer on policy applications
@erdaltekin.bsky.social on the publication process

Plus great presentations throughout💪

#econbluesky

😬

Nothing casual about it. I take these things very seriously.

Finally, I have my Flintstones name.

😂😂

This inspires me to start a collection of pet rock econometricians.

First of all, I added an implementation of extended two-way fixed effects (@jmwooldridge.bsky.social 2021), etwfe(), with inputs and outputs aligned with fetwfe().

There's also now betwfe(), which implements a bridge-penalized (includes lasso and ridge regression) version of etwfe().

👀

🚨 One spot just opened up for our Summer School in Health Economics!

Top-notch lectures by
@jmwooldridge.bsky.social and Stefanie Schurer, presentations, and a great social program are waiting for you.

First come, first served – apply now!

#EconBluesky #economics @healtheconall.bsky.social

It’s a real pleasure teaching a course with @pedrosantanna.bsky.social. Many are calling us “Fire and Ice.”

See you there! Good thinking to hang out in Athens.

To calculate CPI, BLS teams collect 90k price quotes every month covering 200 different item categories.
When data not available, BLS staff typically develop estimates for approximately 10% of cells in the CPI calculation. However, in May, share of data in the CPI that is estimated increased to 30%

I’ve sent responses to editors to pass on to referees after having a paper rejected. But why not let the entire profession — at least those curious— learn from mistakes?

I’d like to see journals allow responses to referees to be published for rejected papers. Sometimes referees are confidently wrong. When a paper is rejected on bad advice, most of the time the referee has no way of knowing they messed up — and might make the same mistake in the future.

I see econometrics in your future.

You can do this in Stata with jwdid if you want the proportionate effects. For levels, you need to do it “by hand.” I assume R is similar.

These things were monoliths. I can see several guys up there. One of them looks familiar, and is clearly just hanging out. He sees me staring at him, and gives a nod. It was Fran Sheehan, the bass player for Boston. A nod from a member of my favorite band when I was 18.

This was not my most minor interaction; we actually spoke and joked. Most minor is probably when I went to see Boston, with Sammy Hagar and another opening act, in Oakland in 1978. In between the first two sets, I was bored, looking around. Behind me about 30 rows back was the mix/sound station.