WGGRAFY / sarla

State space autoregressive length-at-age model
0 stars 1 forks source link

Model fixes #3

Closed seananderson closed 3 years ago

seananderson commented 3 years ago
seananderson commented 3 years ago

Note this samples quickly, but I've totally messed up the model... will fix

seananderson commented 3 years ago

I added a more informative sigma_o prior and that dramatically improves the sampling.

I think the initial conditions are still set up differently than in the paper.