Sampler getting stuck around zero


I’m trying to sample from the posterior of a somewhat complex latent variable model and am having some trouble. In particular, the sampler sometimes get stuck very close to zero for one of my model’s parameters. I’m attaching a traceplot and pairs image to demonstrate.

Has anyone else had similar problems? Any ideas on how to fix it?

If you are that confident that gamma_M_R0_scale cannot be close to zero, then your prior choices should effectively preclude that possibility.

1 Like