Jesse Wolfhagen
@uncertainarch.bsky.social
Bayesian Zooarchaeologist, studying animal domestication and philosophical questions like "how many cows were in this pit?"
GitHub: https://github.com/wolfhagenj
OSF: https://osf.io/va7ye/
GitHub: https://github.com/wolfhagenj
OSF: https://osf.io/va7ye/
...because they're not targeting Serbia?
November 29, 2025 at 6:41 PM
...because they're not targeting Serbia?
Reposted by Jesse Wolfhagen
The aim should be for individual, not institutional, accountability.
Did a university obviously violate the first amendment rights of its faculty to suck up to Trump? Don't seek to fine the university, *seek to jail its president and trustees.*
People make decisions, people bear consequences.
Did a university obviously violate the first amendment rights of its faculty to suck up to Trump? Don't seek to fine the university, *seek to jail its president and trustees.*
People make decisions, people bear consequences.
November 28, 2025 at 6:14 PM
The aim should be for individual, not institutional, accountability.
Did a university obviously violate the first amendment rights of its faculty to suck up to Trump? Don't seek to fine the university, *seek to jail its president and trustees.*
People make decisions, people bear consequences.
Did a university obviously violate the first amendment rights of its faculty to suck up to Trump? Don't seek to fine the university, *seek to jail its president and trustees.*
People make decisions, people bear consequences.
Ha! It's more why I can't get this thing finished. I'll still subject you to an advance copy, for your sins
November 27, 2025 at 2:28 PM
Ha! It's more why I can't get this thing finished. I'll still subject you to an advance copy, for your sins
Also @zenobiewan.bsky.social these are the shenanigans I do instead of just sending you a draft of the piece, I'm sorry
November 27, 2025 at 5:11 AM
Also @zenobiewan.bsky.social these are the shenanigans I do instead of just sending you a draft of the piece, I'm sorry
The scary thing is, the toy model was based on some of my published work! When fitting the model with data, I wasn't seeing the problems. Gabry, et al (2019) (academic.oup.com/jrsssa/artic...) really convinced me into doing these!
Prior predictive checks! You know more than you think you do!
Prior predictive checks! You know more than you think you do!
Visualization in Bayesian Workflow
Abstract. Bayesian data analysis is about more than just computing a posterior distribution, and Bayesian visualization is about more than trace plots of M
academic.oup.com
November 27, 2025 at 5:05 AM
The scary thing is, the toy model was based on some of my published work! When fitting the model with data, I wasn't seeing the problems. Gabry, et al (2019) (academic.oup.com/jrsssa/artic...) really convinced me into doing these!
Prior predictive checks! You know more than you think you do!
Prior predictive checks! You know more than you think you do!
The major culprit turned out to be my "sigma_element" parameter (governing how much my hyper-parameters could vary between different elements), which was left with a "standard" half-normal(0, 1). Looks fine when you type it but is WILD for the scale of my analysis (usually +/- 0.2)
November 27, 2025 at 4:54 AM
The major culprit turned out to be my "sigma_element" parameter (governing how much my hyper-parameters could vary between different elements), which was left with a "standard" half-normal(0, 1). Looks fine when you type it but is WILD for the scale of my analysis (usually +/- 0.2)