Christopher Peters
statwonk.bsky.social
Christopher Peters
@statwonk.bsky.social

Dad, Partner. Statistician-Econometrician-Engineer. AI Developer. Decision theory. Game theory. #autisticpride #9 at Zapier, first DS

Political science 19%
Engineering 18%

Reposted by Christopher Peters

"These mass layoff announcements have not been representative of what most companies in the United States are doing; they're outliers that get a lot of attention... maybe finally other firms start following their footstep, but so far it has not happened."

www.businessinsider.com/amazon-ups-l...
Layoff panic is hiding a surprising reality about the job market
It will take a lot more Amazon-sized layoffs for the job market to actually take a hit.
www.businessinsider.com

Reposted by Christopher Peters

“Lay-offs get a lot of the attention, but weak hiring has been much more important in explaining the cooling that we’ve seen in the labour market.”

Thanks to Stephanie Stacey at @financialtimes.com for chatting w/me about all those scary layoff headlines!

www.ft.com/content/fabb...
Corporate job losses mount amid tightening economy and AI growth
Recent cuts mark end of job security for high-earning knowledge workers whose ranks grew since pandemic
www.ft.com

Reposted by Christopher Peters

Since were enduring another round of Challenger discourse, a reminder -

It’s has some directional signal (when it goes up, so do superior layoff indicators) but the magnitudes are wildly off. Look at those y axes!

(No access to computer so chart not updated)

Reposted by Christopher Peters

CNBC @cnbc.com · 5d
FAA to cut flights by 10% at 40 major airports due to government shutdown
FAA to cut flights by 10% at 40 major airports due to government shutdown
Transportation Secretary Sean Duffy announced Wednesday afternoon that he will be reducing flight capacity by 10% at 40 major airports.
cnb.cx

> President Donald Trump’s disapproval rating has reached an all time high, according to a new poll from CNN.

> About 63% of voters disapprove of Trump’s job as president, numerically reaching the highest disapproval rating across both of his terms in CNN polling.

www.masslive.com/politics/202...
Latest Donald Trump polls: Disapproval worse than after Jan. 6. Here’s why
A new poll has found that President Donald Trump's disapproval rating has reached its highest point across both his terms.
www.masslive.com

"I only want intervals" - said no one ever
Cue about a dozen more think pieces about how the Democrats are out of touch and elitist.
Rep. Clay Higgins (R-LA.), 64, hit out at SNAP users, saying those without a month’s supply of food in their pantries didn’t deserve the help, “stop smoking crack”
www.thedailybeast.com/maga-rep-hig...

Reposted by Christopher Peters

Arctic sea-ice extent currently ranks 5th lowest for the date (JAXA data) since 1989.

Will there be a record low maximum in a few months? The Climate 8-ball quoted Horace, "Great effort is required to arrest decay."

For slowly changing things, I think typically yes (bc wikipedia-representative data is in there). For recent temporalities, it can typically fetch exogenous data.

The cases where neither is true, I find it’s bc the data is strictly private (eg- some financial derivative data).

Reposted by Christopher Peters

sometimes I think math is a whole extremely-long running inside joke that's not actually very funny
You know it’s serious when corporates begin to complain.

$DAL
FL is moving forward w/ plan to end all childhood vaccine mandates. Starting with hepatitis B, chickenpox, and the bacteria causing meningitis and pneumonia. Then next year GOP FL legislature is expected to revisit 1977 law re: whooping cough, measles, polio, rubella, mumps, diphtheria, and tetanus.

Plus, the models can collect and use data. That’s another major source of information for reducing surprise.

Personally I identify a lot of errors, and I’m sure others do as well. So I think the information from people has a strong effect on identifying the errors (the idea people supply the judgement and the models follow instructions).

The models are retrained on their errors. Granted that is currently a slow process today, but many errors are reduced with each update.

Practically they’re all statistical models, and we can evaluate the utility they produce. I think it’s reasonable to include interpretable linear models in the set.

My point being that it would take me a lot longer, and in my mind that’s pretty intelligent (and one example). Plus, it’s pretty easy to identify things they (practically) understand relative to certain human populations.

AI is short-hand for a statistical model, so admittedly I don’t find that answer controversial.

eg- we could have people respond a la lmarena.ai/leaderboard, and compare.

That said, GPT and Claude can write and upgrade a skew generalized t implementation for me (even if yes they make mistakes).
Overview Leaderboard | LMArena
Compare and explore top-performing models across different benchmarks.
lmarena.ai

AI’s smarter than all of us, and gee what an effect on operating leverage / net income?

I think AI’s smarter than all of us, especially rapture people. :)

> An ex-Intel CEO’s mission to build a Christian AI: ‘hasten the coming of Christ’s return’

Intel’s stock is up ~ 70% since this man retired from being CEO.

www.theguardian.com/technology/2...
An ex-Intel CEO’s mission to build a Christian AI: ‘hasten the coming of Christ’s return’
Patrick Gelsinger, executive chairman of Gloo, has made it his mission to advance Christian principles in Silicon Valley
www.theguardian.com

Reposted by Christopher Peters

$SPX is plus 43 basis points, 41 of it is NVDA
I built a DAG diagram with garden hoses for teaching.
Pictured: a collider bias diagram, inspired by a blocked pipe situation I experienced (which I credit with giving me the intuition though it also ruined my belongings in the flooded cellar).
The Walton family (who own Walmart) are worth over $400 billion, yet many of their employees are on SNAP.

Bezos is worth over $400 billion, many Amazon employees require SNAP.

The people who need help are not the problem.

It’s corporate greed. It’s an unwillingness to pay a living wage.

Reposted by Christopher Peters

"You and I could try and talk about this like it were economics or like it were international relations, but it's not. It's a soap opera. It's Sons and Daughters. It's Home and Away. It's Neighbours. And it's going to be a soap opera that goes on and on for the next three and a half years."

It’s best to begin by thinking about causal relationships or approximate them using AI (the process of identifying causes and their effects).

The potential for utility depends on the quality and quantity of our identifications (relative to our current understanding).

Yes, any entity that leads a large scale genocide might lose their money. Not super controversial.

Obviously, Europe and the US should give Russia’s $230 billion in foreign reserves to Ukraine.

Nobody’s going to blame them for table stakes.

Reposted by Christopher Peters

"I can't recall seeing a global heatwave on this scale before.

Forecasts continue to show global average temperatures spiking thanks to heatwaves (relative to average) across the NH & Antarctica.

It's likely average global temperatures will set new records."

(Met4Cast)

#collapse
When there is a random way to do something, there is a less random way that is better but requires more thought. In this case, regression models that make no sense don't belong in a multiverse analysis. An inferential regression without a causal justification is like an opinion without reasons.