AeMCMC is a Python library that automates the construction of samplers for Aesara graphs representing statistical models.
39
stars
11
forks
source link
`normal_horseshoe_finder` does not return the expected posterior. #50
Closed
rlouf closed 2 years ago
We expect the function to return a step function that generates posterior samples for $\tau$ and $\lambda$. However, it calls
horseshoe_posterior:https://github.com/aesara-devs/aemcmc/blob/213e0fd105f57b3f41eb1423d966962512c25ec3/aemcmc/gibbs.py#L275)
which updates the values of $1/ \lambda$ and $1 / \tau$.