search

tjthurman / BAHZ

An R package for Bayesian analysis of hybrid zones
GNU General Public License v3.0
3 stars 1 forks source link

Center the center parameter at 0 for estimation? #18

Closed tjthurman closed 5 years ago

tjthurman commented 6 years ago

Might ease initialization. Could add the mean value back to the results in the chain, to preserve the user-supplied values. Need to see if Stan would work with the center = 0 check removed.

tjthurman commented 6 years ago

Stan works with the center > 0 check removed just fine, and seems to start easily. Need to do some tests, but in this case I also don't think it will matter for the prior if we used the same value for the SD.

tjthurman commented 5 years ago

Did lots of testing, and centering it before loading and then uncentering-it in the cline summary function are effective. However, goes against the philosophy a little bit: I want users to supply a prior for center, not just rely on a broad normal prior centered at the middle of their cline automatically (if they want to do that, they can specify that, instead of it being a default.