Timeline for Using poisson regression for continuous data?
Current License: CC BY-SA 3.0
11 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Jul 14, 2015 at 19:46 | history | suggested | Aaron Hall | CC BY-SA 3.0 |
remove extraneous cruft
|
| Jul 14, 2015 at 19:40 | review | Suggested edits | |||
| S Jul 14, 2015 at 19:46 | |||||
| Jun 12, 2015 at 16:40 | comment | added | RobertF | @PeterFlom - I wonder if this issue comes up a lot because the glmnet package in R doesn't support either the Gamma family or Gaussian family with a log link function. However, because glmnet is used as a predictive modeling package (hence users are only interested in model coefficients, not coeff. stnd errors) and since the Poisson dbn produces consistent coeff. estimates for models of the form ln[E(y)]=beta0 + beta*X with continuous responses regardless of the distribution, I'm guessing the authors of glmnet didn't bother including these additional families. | |
| Oct 19, 2011 at 13:30 | vote | accept | user3136 | ||
| Feb 11, 2011 at 5:22 | history | tweeted | twitter.com/#!/StackStats/status/35931619969736704 | ||
| Feb 11, 2011 at 1:29 | answer | added | Hong Ooi | timeline score: 16 | |
| Feb 10, 2011 at 22:49 | comment | added | Peter Flom | Wouldn't there be other distributions that are better, e.g. whuber's suggestion of gamma? | |
| Feb 10, 2011 at 16:50 | comment | added | user3136 | I have used the gamma distribution for these data. If you use the gamma distribution with a log link you get almost the exact same result you get from an over-dispersed poisson model.However, in most of the statistical packages I am familiar with poisson regression is simpler and much more flexible. | |
| Feb 10, 2011 at 15:15 | comment | added | whuber♦ | How do your empirical distributions differ from Gamma variates, then? | |
| Feb 10, 2011 at 15:15 | answer | added | Simon Byrne | timeline score: 12 | |
| Feb 10, 2011 at 14:59 | history | asked | user3136 | CC BY-SA 2.5 |