Lightnews — Scholar-powered news

Jordan Nafa

@jordannafa.bsky.social

1.3K followers 310 following 2.2K posts

Bayesian Statistician and Data Scientist in the Gaming/Entertainment Industry | Bayesian Statistics, Causal Inference, R, Python, Stan, Decision Theory, Guitar | Former Political Scientist

Posts Media Videos Starter Packs

Pinned

Jordan Nafa @jordannafa.bsky.social · Aug 20

Are you interested in Bayesian Optimization? Do you think multi-arm bandits are cool? Do you like donating to good causes? We'll, I'm super excited to say I'll be hosting a charity workshop on Bayesian Optimization with Multi-Arm Bandits on October 23rd.

sites.google.com/view/dariia-...

Dariia Mykhailyshyna - Workshops for Ukraine

Feedback on the past workshops (if you want to learn how to make wordclouds, check out Text Data Analysis workshop below)

sites.google.com

1 5 19

Jordan Nafa @jordannafa.bsky.social · 45m

A rejection after 18 months is just straight up editorial malpractice 😬

Jordan Nafa @jordannafa.bsky.social · 4h

I had to finally make a spreadsheet with all my inflows/expenses to prove to myself that I could afford it. Basically the same as a new Audi/BMW/Mercedes sedan would be in the U.S, only way more fun to drive.

Jordan Nafa @jordannafa.bsky.social · 6h

TIL, misophonia is the term for when certain prolonged pitches/noises make you irrationally upset

Jordan Nafa @jordannafa.bsky.social · 14h

Update, my wife's response was "Oh my god Jordan, what did you do?" but I'm still married as of writing this.

Jordan Nafa @jordannafa.bsky.social · 15h

So, new car day

1 2

Jordan Nafa @jordannafa.bsky.social · 14h

Jordan Nafa @jordannafa.bsky.social · 15h

It has four different exhaust settings so you can choose how many of its 500 horses you want to subject your neighbors to

Jordan Nafa @jordannafa.bsky.social · 15h

So, new car day

1 4

Jordan Nafa @jordannafa.bsky.social · 17h

Right, I was just here for a regular GT and somehow I overshot that by several orders of magnitude 😮‍💨

Jordan Nafa @jordannafa.bsky.social · 18h

My mother is by far the worst influence in my life

2 6

Jordan Nafa @jordannafa.bsky.social · 21h

I'm not saying Trump v. United States will go down in history as the Dred Scott v Sandford of the 21st century, but I'd be very surprised if it doesn't.

Jordan Nafa @jordannafa.bsky.social · 21h

2010 according to Wikipedia, so only 15 years en.wikipedia.org/wiki/John_Th...

Jordan Nafa @jordannafa.bsky.social · 21h

That picture of Thune is at least 20 years old 😬

2 1

Jordan Nafa @jordannafa.bsky.social · 22h

This will mean killing the Senate filibuster and passing legislation on party-line votes. Any strategy that assumes Democrats gain 60 seats in the Senate is utter fantasy and the person proposing it should be laughed out of politics

Jordan Nafa @jordannafa.bsky.social · 22h

Until the Democratic party is willing to play constitutional hardball and pass meaningful, substantive reforms including supreme court rebalancing/reform, anti-corruption legislation, and substantial constraints on the imperial presidency authoritarianism is here to stay.

1 2

Jordan Nafa @jordannafa.bsky.social · 22h

You can't just keep saying "we'll finally do what we've been saying we'll do for the last decade if we just win a few more elections" forever. At some point your voters will figure out you're just gaslighting them.

Jordan Nafa @jordannafa.bsky.social · 23h

Winning elections is just a means to an end. Ultimately, it's not about winning the election it's about what you do once you have power and if you do nothing to prevent future authoritarianism when you hold power, you get authoritarianism the moment you're out of power.

Jake Grumbach @jakemgrumbach.bsky.social · 1d

“Win every election” is not a viable strategy for preventing authoritarianism

1 1

Jordan Nafa @jordannafa.bsky.social · 23h

All that is to say, it's not enough to simply win elections--you have to be willing to take concrete actions and make meaningful reforms, even if that means tossing out certain rules and institutional norms to do it.

Jordan Nafa @jordannafa.bsky.social · 23h

Jake Grumbach @jakemgrumbach.bsky.social · 1d

“Win every election” is not a viable strategy for preventing authoritarianism

1 3

Jordan Nafa @jordannafa.bsky.social · 1d

So the outcome here is some kind of desirability scale and chocolate and poop are the treatment conditions? So

E[E[Y(Chocolate)] - E[Y(Poop)]]

1 1

Jordan Nafa @jordannafa.bsky.social · 1d

So, I think part of the problem is that learning requires some degree of curiosity and a willingness to explore beyond the strict bounds of the course. In my experience, this isn't expected as much of undergrads as it is of grad students but ultimately, if you lack that curiosity you won't gain much

1 2

Jordan Nafa @jordannafa.bsky.social · 1d

You have to dedicate like a whole week to "this is what a folder is," "this is how you find a file path," "you must replace all \ with / because \ is an escape character"...

1 4

Jordan Nafa @jordannafa.bsky.social · 1d

The most painful realization back in my teaching days was that no one under 30 understands what a file system is anymore because Microsoft and Apple have both tried to obfuscate it so heavily over the last 10-15 years.

1 5

Jordan Nafa @jordannafa.bsky.social · 1d

I'm not even sure what the estimand here is...

Ian Hussey @ianhussey.mmmdata.io · 1d

Make an effect size prediction!

@jamiecummins.bsky.social and I are replicating Balcetis & Dunning's (2010) "chocolate is more desirable than poop" (Cohen's d = 4.52)

Let us known in the replies what effect size you think we'll find. Details of the study in the thread below.

1 2

Jordan Nafa @jordannafa.bsky.social · 1d

Dogs are loving chilling in the sun on the back patio this morning

Jordan Nafa @jordannafa.bsky.social · 1d

The moral of this story is that you should never attempt to measure time by counting things. Just diff the minimum and maximum timestamp instead

Jordan Nafa @jordannafa.bsky.social · 1d

Event*, not eventually lol