DO. NOT. LISTEN. TO. THIS. MAN. ON. CLIMATE. newsweek.com/right-path-for…

Quant methods hot take: If you lack the domain knowledge required to draw the relevant DAG or set meaningful priors, you should neither fit nor attempt to interpret the model. Back away slowly from your stats software and do a comprehensive lit review.

So I just fit a model 1st try with 37 Gaussian Processes (variably supported) nested in a GP. >7K true params, init=NULL, warmup=1000, iter=1000. Fit: finite time (8hr), 0 divergences, max(Rhat)=1.009, min(ESS)=444 (4 chains). 🙌 Impossible before this course: 👏👏👏 Stan

Every now & then I question my life choices. Currently knackered, stinking of putrid calving, hungry & an hour away from a shower. Today is one of those days. #StinkingPlusHungry #CowVet

Why don't experts agree? They're often interpreting research with different outcomes in mind. To remain the "expert" (in the public's eyes) they risk doubling down rather than seeking reasons for their differences. *We* created the environment where an expert can't be wrong 😔

The biggest story in microbiology since they renamed Pasteurella trehalosi. #bluesoup

In 2016 Luis Alberto Quiroga Cairo produced the Datasaurus Dozen to make this point even more dramatically. It’s absolutely wild that 50 years later we still need to communicate this to folks who have recently awarded STEM degrees. Dataviz needs to be a fundamental part of data education

Loving this version! #dataviz

Coffee & feet up. You need to watch this today. youtu.be/DOWDNBu9DkU

Sound familiar to any other cow people out there? Potentially great though 👍 youtu.be/3sUug1kfNps via YouTube

Why is it that the bigger the claimed effect size the more excited people become? Surely for large effects it should be the opposite? Just read a 391 page PhD backing a recent milk yield effect claim. Suddenly feel quite hollow about the way it's been communicated to/by others.

Watching one side of my Twitter life insulting big chunks of the other. #hesnotwrong

Magical evening at the Minack. Great night!

Effect of controlling on a mediator? Exercise -> muscle % -> energy expenditure/KgBWT Proposed corrected interpretation: "little else except body composition affects energy expenditure for given exercise?" Richard McElreath 🐈‍⬛

Love the crisis of interpretation these slides have caused. Great example to expose how confusing statistical "adjustment" can be. Adjusting for being tired, people who sleep more hours are no more aware than those who sleep less.

That moment when you wake up at 2am with a sudden revelation about something you *thought* you deeply understood for a decade, & being 100% certain you're only the 2nd person ever to achieve #21dPR zen. #Eicker 🧘

Should our meat & dairy come from 'we use no drugs of any kind' farms, or should farm animals have access to modern healthcare & pain relief, provided medicines are used responsibly & with food safety as a priority?