Jim Savage

It would be better if really nice things were not so damned expensive

Oh god. I need to get my arsenic levels checked stat.

Why are panpipes?

Start watching a bunch of really weird films. Or better, really great, under-appreciated ones. Like Sorcerer.

a portrait of Tenochtitlan a 3D reconstruction of the capital of the Aztec Empire

If you’ve ever spent even one day in Mexico City, don’t miss this.

tenochtitlan.thomaskole.nl/index.html

100% savory!

I genuinely was amazed it’s not more of a thing

A nearby restaurant has the following appetizer on their menu: donuts drizzled with truffle oil, covered in pecorino and parsley, served with marinara sauce. $10.

from news.gallup.com/opinion/poll...

nice context for some of the progress the country has made over the last 80-odd years on racial/ethnic prejudice.

It’s so good.

The 10yo is obsessed with Seinfeld. And, strangely, Eraserhead.

Yep, and a couple of orders of magnitude faster (effectively independent draws from the posterior per unit of time) than what it replaced!

Have you put it on a preprint server yet? It wonder if you put the study design on the social science prediction platform, get a bunch of people to predict some nonzero effect, then include in your paper that a null result is interesting and unexpected

🛰️ #envecon

So one of our key findings was the size of environmental damages from satellite constellations determines whether the market will make them too big or too small relative to what's socially optimal

Some evidence on those damages here: https://www.pnas.org/doi/full/10.1073/pnas.2313374120

Github diff showing addition of code: def _get_change()-> Decimal: daily_volume = current_session().query(func.sum(Trade.size * Trade.price)).filter( Trade.created_at > datetime.now() - timedelta(days=1)).scalar() or Decimal() return f2d(numpy.random.normal(7500, 3000))* daily_volume / Decimal('1e9') @always_run_in_transaction(ro=False) def update_public_insurance_fund(): change = _get_change() sess = current_session() + public_insurance_fund = Public InsuranceFund.get(sess) sess.add(Public InsuranceFundChange(public_insurance_fund-public_insurance_fund, size-change)) public_insurance_fund.size += change sess.commit()

FTX @FTX_Official The 5.25 million FTT we put in our insurance fund in 2019 now makes the fund worth over 100 million USD https://ftx.com/status Screenshot: Backstop Fund Size: 5,478,274.51613972 USD, 5,250,000.000000000 FTT Last updated: 14/02/2021, 08:05:00

From yesterday's exhibits in US v. Sam Bankman-Fried:

The prosecution shows that the "insurance fund" that FTX bragged about was fake, and just calculated by multiplying daily trading volume by a random number around 7500

I've developed a new forecasting tool! You tell me what your education intervention is, including the full details of how it changes teacher and student behaviors. Then I use advanced AI methods to predict that the effect size is 0.1 SDs.

#econsky

Fun undergraduate econometrics exam essay in here somewhere

Fun story: went to Cuba in 2010. The only evidence there that the US existed was Tabasco and Inglourious Basterds, which somehow escaped the censors.

Inglourious Basterds?? The Big Short?

Skip leg day? Worry not.

Who do I know in Boston area who’s game for a dinner party next week?

Only the best red wine with yellow plastic lid from the Dominican Republic for my family.

Carol Greider - Same day, a Nobel prize and a grant rejection. Carol Greider - Same day, a Nobel prize and a grant rejection!

Re: recent discussion about Katalin Kariko--
In 2009, Carol Grieder won the Nobel Prize for her work on telomeres. That same day, she got a notification that her NIH grant proposal was deemed not worthy of discussion due to "limited preliminary data."

The fact Philly is not a coastal city and Chicago is just feels wrong.

The fact Philly is not a coastal city and Chicago is just feels wrong.

Metascience in one tweet

Hearing the two stories about Francesca Gino and Katalin Kariko within a short window of time is insane.

Me trying to remember the name of the RA that did all analyses of a seminal paper that got me a $500,000 salary.

Currently downloading 200k papers. Doing research with network data is fun. Accessing it is a pain.