Matt Hauer

No. The algorithm DP is very useful. I want to be angry at the algorithm because of noise in census data, but the algorithm itself is stupidly useful.

I’m actually kind of pissed at how useful differential privacy is.

IYKYK

Straight killing the pinewood derby today! 🙌👑

Context is king

Pretty much, yep.

“These days are going to pass like a grass fire/ so don’t let anybody tell you that you’re safe”

Thanks Valerie!

Thanks! Yea the winning paper looks incredible!

Thank you!

Thank you!🙏

Sadly I don’t have a Bluesky thread over here and I’m a little loath to link the thread from the other place.

Thank you Mary! 🙏

PNAS Announces Six 2024 Cozzarelli Prize Recipients - NAS WASHINGTON, DC – The Editorial Board of the Proceedings of the National Academy of Sciences (PNAS) has selected six papers published by PNAS in 2024 to receive the Cozzarelli Prize, an award that reco...

My paper on Climate Migration and Demographic Amplification was selected as the Finalist for the Cozzarelli Prize! (www.nasonline.org/news/2024-co...)

It’s really bad everywhere. I know I’m not the only one looking for greener pastures.

I’m not sure AI is a bubble. It’s bubble-like in that there’s lots of snake oil but the underlying core tech is here to stay. We can talk more about it at PAA.

Now is probably the best time to consider leaving the academy. Sadly, it’s probably only going to get worse from here.

Just casually browsing Reddit when you see your study.

Haha!

The understated athleticism of the ninjabread man, the tasteful googly eyes, the artful smattering of sprinkles. Masterpiece.

This might be my favorite gingerbread man we decorated.

Not a bad 2024. 6 papers, including 1 in PNAS and 1 in Demography with @valmuellerasu.bsky.social!

So the meme is both right and wrong and it is incumbent upon us to ensure that we are on a low emissions pathway, ensuring that older generations have peak lifetime carbon emissions.

We also decompose the contributions changes in population size and age-specific emissions rate in changing cohort emissions. It is the age-specific emissions that drive most of the changes in cohort emissions rather than cohort sizes.

The cohorts with the highest life time emissions depends on which emissions pathway we are on. Generally, higher emissions scenarios = younger cohorts with higher lifetime emissions and lower emissions scenarios = older cohorts.

So we create a demographic version of both IPAT and the Kaya Identity that allows us to estimate carbon emissions by cohort, taking into account changes in carbon emissions across time and, imo importantly, over the life course.

On one hand, its true that older cohorts have clearly had more time to emit CO2 emissions than younger generations. On the other hand, younger generations could be emitting more over their lifetime due to higher baseline, societal emissions.

I came across this meme while browsing on Reddit about a year ago and after chuckling about it for a moment I wondered if it were true.

Using formal demographic methods, we develop an approach to estimate carbon emissions across countries, time periods, and, importantly, birth cohorts. And we then estimate the total lifetime carbon emissions of each birth cohort for each country since 1850.