Timeline for Which distributions have closed-form solutions for maximum likelihood estimation?
Current License: CC BY-SA 3.0
5 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 12, 2012 at 1:58 | history | edited | Michael R. Chernick | CC BY-SA 3.0 |
added 6 characters in body
|
| Jul 12, 2012 at 1:57 | comment | added | Michael R. Chernick | Thnaks Neil for pointing that out. I guess not all exponential family distributions have closed form solutions. | |
| Jul 12, 2012 at 1:28 | comment | added | Neil G | It's not always a "nice function", for example, the sufficient statistic of the beta distribution is $[\log x\; \log (1-x)]^{\rm T}$, from which numerical methods are required to find the shape parameters $a$ and $b$. | |
| Jul 11, 2012 at 14:14 | comment | added | Macro | This question is incredibly broad but it appears the OP may be asking what characterizes a distribution that has a closed-form solution for the MLE rather than asking for an exhaustive list. In any case, an exhaustive list isn't even possible. | |
| Jul 11, 2012 at 14:01 | history | answered | Michael R. Chernick | CC BY-SA 3.0 |