Timeline for Eliminating divergent transitions in Stan
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 5, 2021 at 0:43 | comment | added | desert_ranger | Never mind! I got it. Thank you once again. | |
| Mar 5, 2021 at 0:32 | comment | added | desert_ranger | Thank you. Could you please elaborate just a little more. For instance, in a uniform prior, I would be assuming all values between a certain range. What exactly does Cauchy do in this case? Feel free to suggest some reading material :) | |
| Mar 5, 2021 at 0:30 | comment | added | Demetri Pananos | @desert_ranger Long and the short of it is that it allows the parameter to be big if it needs to be, but has most of the mass near 0. It isn't a good default for everything, but it appears to be better than the uniform prior in a lot of cases. | |
| Mar 5, 2021 at 0:22 | comment | added | desert_ranger | Can you please tell what was the rationale behind choosing a Cauchy prior? How does that help reduce the divergences? | |
| Mar 4, 2021 at 22:34 | vote | accept | desert_ranger | ||
| Mar 4, 2021 at 22:12 | history | answered | Demetri Pananos | CC BY-SA 4.0 |