Timeline for Can you specify correlated coefficients in Stan models?
Current License: CC BY-SA 4.0
4 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 23 at 4:10 | answer | added | Demetri Pananos | timeline score: 2 | |
| Mar 22 at 17:29 | comment | added | Corned Beef Hash Map | That sounds exactly like what I want. Do you have to specify any new parameters for that? Or can you specify a joint prior? I tried looking at the documentation for "Multivariate Changes of Variables" and didn't understand what it was talking about. Am I looking in the wrong place? Thanks for your help, I'm pretty new to Bayesian stats in case that's not obvious haha. | |
| Mar 22 at 5:28 | comment | added | Demetri Pananos | Why not put a prior on the joint distribution of $\alpha$ and $\beta$? That way, you can specify the correlation be whatever you want | |
| Mar 22 at 3:31 | history | asked | Corned Beef Hash Map | CC BY-SA 4.0 |