iiinova — where your digital self keeps evolving

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 11/04/2026

Yes I know but I’m in full day white paper committee meetings. :/

210

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 11/04/2026

Argh why am I always to late for relevant conferences! Looks good!

120

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 11/04/2026

Ah of course, that’s the one. Thanks for the praise, happy mtefe is useful. I dunno, these days any AI would probably do a fair translation of mtefe and mtefeplot from Stats to R?

000

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 11/04/2026

Didn’t Alex torgovitsky make a fancy MTE package for R with partial identification and stuff?

100

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 19/12/2025

We did a thing! Cool project with amazing coauthors, see Luisa's summary and the NBER WP below!

032

Reposted by Martin Eckhoff Andresen

Askill Harkjerr Halse @halsevesenet.bsky.social · 28/05/2025

Det er mye forvirring rundt begrepet akademisk frihet om dagen. Dette synes jeg var en ganske bra opprydding. www.dn.no/kommentar/gl...

dn.no

Retten til elendig forskning (+)

Akademikere roper opp om Glenn Diesens (selvfølgelige) rett til å mene hva han vil. De bør heller plukke fra hverandre forskningen hans.

051

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 16/05/2025

Wow. What do you do next? Nothing? I conjecture they might have a harder time recruiting reviewers in the future…

160

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 07/05/2025

In this paper, we exploit the bunching in income statistics in Victor Orban’s Hungary to back out the political cost of manipulating official statistics.

052

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 27/03/2025

My paper with @sturla.bsky.social on marginal students and high school dropout is out in JOLE!

070

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 27/02/2025

Good idea - do you have a good example?

000

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 27/02/2025

Gotta admire all the work that went into this, despite the sad findings. Thanks, I4R!

0192

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 27/02/2025

require more scrutiny of mechanisme and possible explanations for why the treatment effects can be so large. Doesn't always feel fair that either, but better than any alternative I've found.

130

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 27/02/2025

my thoughts exactly. It's hard be skeptical as a referee thought without hanging ut on some specific methodological shorthcoming or explanation, without outright accusing the authors of fraud? When I find implausibly large effect sizes, but has no methods issues to hang it on, my go-to comment is to

110

Reposted by Martin Eckhoff Andresen

Steve Cicala @stevecicala.bsky.social · 27/01/2025

The @economist.com gets it. Norway closing itself off from the European grid is like Florida prohibiting citrus exports. www.economist.com/leaders/2025...

economist.com

To make electricity cheaper and greener, connect the world’s grids

Less than 3% of the world’s power is internationally traded—a huge wasted opportunity

0134

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 20/01/2025

But it seems to work when my command sets e(cmdname) instead…

100

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 20/01/2025

I don’t know exactly why, but if your program sets e(cmd), bootstrap will throw an error after the final bootstrap repetition has ended, during the display of the results. Had to simply remove setting e(cmd). Without e(cmd), however, you cannot use estimates store on the results.

100

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 20/01/2025

When you google a technical problem, find a post at Statalist and for a split second believe you have found someone with the exact same problem, only to realise it is you, in 2018. And nobody ever responded.

2120

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 05/01/2025

Sample splitting would work, I guess? Test in half, run either TWFE or saturated estimator in other half, reverse roles?

000

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 03/01/2025

This is a clever idea! Very straightforward to implement. Is there an issue with pre-testing here, e.g. if researchers use this test, and run with TWFE if p>0.05? Similar in spirit to first stage testing with IV.

110

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 17/12/2024

Mandatory pic of outdoors opportunities just 20 minutes from the city centre with the metro:

000

Martin Eckhoff Andresen @eckhoffandresen.bsky.social · 17/12/2024

What Linnea said! Great phd program in wonderful Oslo, fully funded. Happy to answer questions! Deadline January 15th.

130