I was at Google today to give a talk about Chapter 7 of Probably Overthinking It: Causation, Collision, and Confusion.

I'll post the video when it's available, but in the meantime, the slides are here: docs.google.com/presentation/d…

This week's installment of Data Q&A is about testing differences in the 85th percentile

allendowney.com/blog/2024/04/2…

Different models yield different p-values, but that's ok -- they don't have to be precise.

In 1990, over 9% of the world's newborns did not survive beyond age five.

By 2021, this child mortality rate had declined to 3.7%.

Thanks to progress in global health, 132 million more people are alive in our world today.

New article by Max Roser → ourworldindata.org/children-saved…

A recent review of Probably Overthinking It:

sekarwrites.com/review-and-sum…

This is the topic of Chapter 8 of Probably Overthinking It and this talk at SciPy 2023: youtu.be/-rE3DfeZ_jE

Superflares and CMEs are my examples of Black Swan events that might be untameable.

The latest installment in the Data Q&A series is about estimating percentiles, the limits of bootstrapping, and quantifying uncertainty due to missing data.

allendowney.com/blog/2024/04/2…

Hey! This is exactly the topic of my talk earlier today!

I have arrived at ODSC East 2024, getting ready for my talk: odsc.com/speakers/who-w…

I'll post the slides after the talk.

Which plot indicates a stronger relationship?

Discussion here:
allendowney.com/blog/2024/04/2…

Probably Overthinking It is a staff pick Newtonville Books

What does a confidence interval mean?

The usual explanation is complicated and confusing. In this article, I argue that it can be simple -- and you don't even have to be a Bayesian.

allendowney.com/blog/2024/04/1…

I just dropped a heterodox opinion in the Reddit statistic forum. If you are interested in the philosophy of probability, you might find it interesting.

But I really, really understand if you don't :)

reddit.com/r/statistics/c…

I was just about to argue with someone on Twitter, but at the last minute, I changed my mind.

Instead, have some baby owls.

Based on its writing style, how old does ChatGPT sound to you?

I have an estimate, based on decades of reading student writing, but I'm curious what other people think.

Here's this week's installment in a new series, Data Q&A: Answering the real questions with Python.

'Is taking the SD of a count variable helpful?'

allendowney.com/blog/2024/04/1…

'Pure math is beautiful (and occasionally useful) and applied math is useful (and occasionally beautiful). But there’s no reason to study fake applied math that is neither beautiful or useful.'

Here's the first in a series of blog posts where I use Python (and Jupyter notebooks) to answer questions from Reddit's statistics forum.

'Can you calculate a standard error for harmonic mean?'

Yes!

allendowney.com/blog/2024/04/0…

Here's my answer to a question on Reddit -- I'd be interested to know what people here think...

'What do you think will be the impact of AI on the role of statisticians in the near future?'

reddit.com/r/statistics/c…

I'm getting back to Elements of Data Science after some time, and finding several places where I try to do too much, too fast.

Fortunately, there is a solution to this problem -- CUT! And there are no revisions more satisfying than a good cut.

Since Colab is running PyMC version 5 now, I've updated the MCMC chapter in Think Bayes.

Links to the new (and old) chapters are here: allendowney.github.io/ThinkBayes2/