John T Ormerod

I like likelihoods.

It is a model all the way down.

ChatGPT gave a significant portion of my students the wrong answer to an easy quiz question that everyone used to always get right

I had two groups of students turn in annotated bibliographies that were ChatGPT generated - not a single citation actually existed

I taught this for the first time in my course this year. We ended up doing a couple of weeks. It isn't much, but at least they got a taste of it.

GitHub - MilesMcBain/more: Serve up more when things go bad Serve up more when things go bad. Contribute to MilesMcBain/more development by creating an account on GitHub.

Make errors more accessible and better with #rstats package {more} by @milesmcbain.bsky.social - inspired by rustc's detailed error explanations! This idea feels so good.

github.com/milesmcbain/...

What an interesting sounding course!

#GazaGenocide #MarchForHumanity #AlboAndWongAreFucked #ChrisMinnsIsFucked

As my mum would say: :Butter wouldn't melt in its mouth".

I make the sun rise by waking up.

TSGTA =Three statisticians gave ten answers.

New #rstats blog up!

solomonkurz.netlify.app/blog/2025-07...

This is the second post in the series where we use {brms} to learn {Stan} code. This time we review matrix notation, model matrices, and a new class of distribution functions in Stan.

two men are standing next to each other in a room . ALT: two men are standing next to each other in a room .

Oorah. The other day I felt real feelings.

Owf. Damage done.

I miss random too, but am all for an adopted kitten, yay.

Fiducial inference

The new normal.

So senseless. What a waste of life.

God loved everyone whoever they are, man, woman or any other child of god.

And what is your evidence against it?

To double down as a statistician, what is your null hypothesis?

Not all fields are at the same levels of maturity. Physics is at a different level of maturity as the social sciences, for example. Do you have an expectant difference in replicability for *any* field for the replicability given that all fields are exploring the forefront of their research areas?

As the prior probability of any arbitrary molecule curing a cancer is extremely low than by Bayes rule alone even an extremely efficient science would produce many false positives.

Frequentist super-raft vs Bayesian superyachts, I know what I am choosing.

Then why use the word crisis? If science isn't stagnating, is there really a crisis?

People might see it as a consequence of a replication crisis.

Bayesian Superyachts are all the craze these days Steph. I put in a grant for one just the other day.

I find all of this really odd. I don't know about other fields, but there has been dramatic advances in cardiovascular and cancer research in recent history, i.e., where a whole heap of money has gone into. Advances in science haven't been uniform, but they are certainly there.

Happy birthday Cameron. Had my fair share. They aren't that fun.

Hope it passes soon, whatever it is.