Lightnews — Scholar-powered news

Matthew Gordon

@mdgordo.bsky.social

980 followers 360 following 38 posts

environment, development, economics. american in paris. AP at @PSEinfo

Posts Media Videos Starter Packs

Matthew Gordon @mdgordo.bsky.social · May 21

A tax on remittances is basically a tax on global poverty reduction. It must be one of the most regressive policies being considered by the current admin, yet I haven't seen much discussion or analysis

www.semafor.com/article/05/1...

House Republicans’ proposed tax on remittances spooks financial sector

“This came totally out of nowhere,” one cryptocurrency executive told Semafor.

www.semafor.com

Matthew Gordon @mdgordo.bsky.social · Apr 10

One of my best students just had his funding offer for a PhD in the US withdrawn. Meanwhile we've had a massive surge in applications to our program from really strong US and international students. It feels like a shift is already underway.

Matthew Gordon @mdgordo.bsky.social · Mar 19

Probably better off creating an index if the goal is dimensionality reduction.

Matthew Gordon @mdgordo.bsky.social · Mar 18

Comments welcome! We are currently working on a more econ-audience version of this paper which will replicate a few existing studies that use remotely sensed dependent variables.

Matthew Gordon @mdgordo.bsky.social · Mar 18

We outline several other ways this type of error can occur in the paper, and propose methods for fixing it - described by @lcsanford.bsky.social in his thread above.

Matthew Gordon @mdgordo.bsky.social · Mar 18

Another example - say you are measuring air pollution using satellite data. We know that satellite measures of pollution saturate at higher concentrations. If your treatment only has an effect in highly polluted areas, the satellite data might detect no change, even if there was an effect.

Matthew Gordon @mdgordo.bsky.social · Mar 18

Imagine that you are measuring deforestation and your ML model sometimes confuses forests for irrigated cropland. If your treatment affects cropland, this might mistakenly look like an increase in forest to your model.

Matthew Gordon @mdgordo.bsky.social · Mar 18

But it is really differential measurement error that matters - errors that are correlated with treatment status. How can this happen, even in an RCT?

Matthew Gordon @mdgordo.bsky.social · Mar 18

One thing that sometimes surprises people, is that, even in an experimental setting, treatment effects can be biased if the dependent variable has measurement error. This is clearly the case if the dependent variable is a prediction from a machine learning model.

Matthew Gordon @mdgordo.bsky.social · Mar 18

Check out our new working paper on using techniques from the algorithmic fairness literature to correct machine learning prediction errors in a causal inference setting.

Luke Sanford @lcsanford.bsky.social · Mar 17

1/9
We are excited to share our new working paper:
arxiv.org/abs/2502.12323
If you use ML predictions (like remote-sensed data) as outcomes, the resulting regression coefficients can be biased by measurement error. With @megan-ayers.bsky.social @mdgordo.bsky.social @eliana-stone.bsky.social

Adversarial Debiasing for Unbiased Parameter Recovery

Advances in machine learning and the increasing availability of high-dimensional data have led to the proliferation of social science research that uses the predictions of machine learning models as p...

Reposted by Matthew Gordon

Luke Sanford @lcsanford.bsky.social · Mar 17

1/9
We are excited to share our new working paper:
arxiv.org/abs/2502.12323
If you use ML predictions (like remote-sensed data) as outcomes, the resulting regression coefficients can be biased by measurement error. With @megan-ayers.bsky.social @mdgordo.bsky.social @eliana-stone.bsky.social

Adversarial Debiasing for Unbiased Parameter Recovery

Advances in machine learning and the increasing availability of high-dimensional data have led to the proliferation of social science research that uses the predictions of machine learning models as p...

Reposted by Matthew Gordon

Zack Beauchamp @zackbeauchamp.bsky.social · Mar 10

A recent report showed that the destruction of just one USAID program, the anti-AIDS PEPFAR initiative, would lead to the deaths of one million people *every year* www.vox.com/future-perfe...

Reposted by Matthew Gordon

Nick Hagerty @hagertynw.bsky.social · Feb 23

This year's Bozeman Applied Micro Conference is June 16-17, 2025

Paper submissions open now through March 7

www.montana.edu/econ/summerc...

Applied Economics Summer Conference - Agricultural Economics and Economics | Montana State University

The Department of Agricultural Economics & Economics at MSU are hosting an economics conference in the summer of 2022.

www.montana.edu

Matthew Gordon @mdgordo.bsky.social · Jan 21

Hosting a workshop on 'Natural Capital' @pse.bsky.social on Feb 7 with a great lineup
@ludogazze.bsky.social @bengroom.bsky.social @floriangrosset.bsky.social and more.

Those interested are welcome to register to attend:
www.parisschoolofeconomics.eu/en/events/na...

Natural Capital

The Paris School of Economics is glad to invite you to the "Natural Capital" workshop organized by the Measurement in Economics Chair.

www.parisschoolofeconomics.eu

Matthew Gordon @mdgordo.bsky.social · Jan 20

not that i know of unfortunately but you can try messaging the organizers

Matthew Gordon @mdgordo.bsky.social · Jan 20

PhD students consider submitting to our annual conference run by and focused on doctoral students in economics. April is a nice time to visit Paris

docs.google.com/forms/d/1u7v...

docs.google.com

Matthew Gordon @mdgordo.bsky.social · Dec 11

Awesome it looks like I am staying close to there

Matthew Gordon @mdgordo.bsky.social · Dec 11

On my way to Copenhagen for the STEG Agriculture conference. First time in town. What should I do?

Matthew Gordon @mdgordo.bsky.social · Nov 29

In 2015 Pakistan was signing 30 year contracts with coal producers. That was the year solar became cheaper than coal.

Matthew Gordon @mdgordo.bsky.social · Nov 29

“Annual capacity payments to private power producers are six times higher than Pakistan’s annual healthcare budget, yet nearly half of the power capacity remains unused.” voxdev.us10.list-manage.com/track/click?...

Nice policy relevant piece from job market candidate Sugandha Srivastav.

Why Pakistan is locked into overpriced and environmentally damaging power sector contracts

Annual capacity payments to private power producers are six times higher than Pakistan’s annual healthcare budget, yet nearly half of the power capacity remains unused. What explains this phenomenon?

voxdev.us10.list-manage.com

Matthew Gordon @mdgordo.bsky.social · Nov 25

Lecture slides for 'Air Pollution and Welfare' this week. I try to use the topic to teach an appreciation for synergies between reduced form and structural methods:
mdgordo.github.io/personalwebs...
Comments welcome!

Matthew Gordon @mdgordo.bsky.social · Nov 25

Assume this leak would have gone unnoticed for a year. Using EPAs social cost of methane, the value of stopping this one leak was 75% of the cost of developing and launching the satellite.

www.prnewswire.com/news-release...

Carbon Mapper Achieves First Tanager-1 Methane Mitigation Success

/PRNewswire/ -- Carbon Mapper released over 300 methane and CO2 plume detections today— its first tranche of emissions data based on observations from the...

www.prnewswire.com

Reposted by Matthew Gordon

Anna Papp @annapappp.bsky.social · Nov 21

I'm on the job market! My #EconJMP examines climate adaptation inequality in labor markets (see ⬇️).

This is part of my research on the effects of climate change/plastic pollution on workers & society - often with unique data & an interdisciplinary lens.

Learn more: https://pappanna.github.io/

Matthew Gordon @mdgordo.bsky.social · Nov 17

PSE is hiring for two positions this year - applications due next week: www.parisschoolofeconomics.eu/en/job-oppor...

Reupping my thread from last year about why Americans should consider applying.

Matthew Gordon @mdgordo.bsky.social · Nov 14

Me!