Timeline for Logistic Regression on very rare events - R
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 30, 2018 at 14:58 | answer | added | AdamO | timeline score: 1 | |
| Jan 30, 2018 at 14:39 | history | edited | AdamO | CC BY-SA 3.0 |
deleted 52 characters in body
|
| Mar 5, 2016 at 22:30 | history | tweeted | twitter.com/StackStats/status/706245601890009089 | ||
| Sep 12, 2013 at 5:26 | comment | added | Alex Holcombe | Brglm has worked for me to address separation. E.g., pic.twitter.com/vOQgwJDL0T You can get brglm's technical background, Kosmides 2007 Bias Reduction in Exponential Family Nonlinear Models by googling it. | |
| Aug 28, 2013 at 14:10 | comment | added | Scortchi♦ |
The phenomenon is called separation & one way of dealing with it is to use penalized likelihood fitting to reduce the bias of maximum likelihood estimates. I don't know what you can do in speedglm.
|
|
| Aug 28, 2013 at 13:39 | history | asked | Davy Kavanagh | CC BY-SA 3.0 |