ᴅᴀɴɪᴇʟ ᴍɪʟʟɪᴍᴇᴛ ✡️☮️❤️

No! Not measurement error! And one-sided at that! There’s a cure for that…

www.cambridge.org/core/journal...

There are 0 ways where Wisconsin fb can score N points ∀ N>0

A perfect surname for one who studies how things change.

www.jstor.org/stable/2340499

On the (Mis) Use of the Fixed Effects Estimator Data that span multiple units and time periods allow controlling for time-invariant heterogeneity correlated with the covariates. While researchers can do this in different ways, the fixed effects es....

"On the (Mis) Use of the Fixed Effects Estimator"

(by @dlmillimet.bsky.social and Marc F. Bellemare)

published in the Oxford Bulletin of Economics & Statistics.

onlinelibrary.wiley.com/doi/10.1111/...

Thanks for publicizing! 🙏

Designed for an introductory course in probability and statistics for economics and business undergraduates, Jason @abrevaya.bsky.social's textbook "Probability and Statistics for Economics and Business" introduces students to the R statistical programming language: mitpress.mit.edu/978026255336...

Agree and applaud you. Unfortunately you’re a rare bird.

I wholly agree. Most wouldn’t bother, but it’s brilliant and extremely useful.

😂

It’s amazing what you can learn from history. Probably why certain people are so adamant about controlling the curriculum starting in kindergarten.

That sounds like so much effort 😂

If I recall the mission statement is spot on for what John, Marc, and I have discussed. Maybe you’ll let us join the board and run a practioners corner?!

Yeah. And the key point we also remind users is that for FE to remove heterogeneity not only must the attribute be time invariant but the return to the attribute must also be. With data that spans the onset of analytics and the ever changing analytics, this clearly has changed what “ability” matters

But … it’s not diff in diff?!?

Thanks, John. Marc and I thought this was an important paper to write. It was an issue that has long bothered both of us. Hopefully it helps applied peeps think a bit more carefully about their model.

Thanks, John! 🙏

Is it a win + noise?

Lurking in the background! No classical measurement error for the win.

lol

You love staggered DID?! You wanna bet your life on it? 😂

👏👏👏

Policy relevance of causal quantities in networks In settings where units' outcomes are affected by others' treatments, there has been a proliferation of ways to quantify effects of treatments on outcomes. Here we describe how many proposed estimands...

When treating one person might affect others, there's an explosion of possible ways to quantify these effects.

Which of these are relevant to applied problems — like choosing a policy governing treatment?

This is the jumping off point for this new short paper:
arxiv.org/abs/2507.14391

As someone who was just in Europe for a conference … GET SOME AC! 😂 it was brutal

More like murder-suicide in the end since tariffs just screw us

“The Party told you to reject the evidence of your eyes and ears. It was their final, most essential command.”

~ George Orwell, 1984

*combined* 😳

Revisiting the Omitted Price Bias in the Estimation of Production Functions The lack of information on quantities presents significant challenges for estimating production functions. As shown by Klette and Griliches in 1996, deflating nominal variables using aggregate price ...

#econsky

A cool and important example of the importance of measurement error and the interesting research that can be done by taking the problem seriously.

doi.org/10.1111/joie...

Sushi 🍣. Easily every day.

lol. Someone needs a lesson on PI.