Timeline for Intuition behind logistic regression
Current License: CC BY-SA 3.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| May 24, 2020 at 3:47 | answer | added | HXD | timeline score: 1 | |
| Sep 29, 2013 at 19:50 | vote | accept | user16168 | ||
| Sep 27, 2013 at 6:08 | history | tweeted | twitter.com/#!/StackStats/status/383473092557045760 | ||
| Sep 26, 2013 at 21:41 | answer | added | Bill | timeline score: 7 | |
| Sep 26, 2013 at 20:57 | comment | added | whuber♦ | A worked illustrated example of (1) where a "sigmoid" is not used appears at stats.stackexchange.com/a/70922. That answer includes an explanation of (2). Another example appears at stats.stackexchange.com/questions/63978/…. A more mundane (but less technical) discussion occurs at stats.stackexchange.com/a/69873, focusing on issue (2). | |
| Sep 26, 2013 at 20:06 | comment | added | gung - Reinstate Monica | @AdamO provides an excellent overview below. If you want more detailed information about what it means that the logit is the 'canonical link function', you may want to read Momo's answer here: difference-between-link-function-and-canonical-link-function-for-glm. | |
| Sep 26, 2013 at 19:46 | answer | added | AdamO | timeline score: 12 | |
| Sep 26, 2013 at 19:36 | comment | added | gung - Reinstate Monica | My answer here: is the logit function always the best for regression modeling of binary data, may also be helpful in thinking about different possibilities. | |
| Sep 26, 2013 at 19:34 | review | First posts | |||
| Sep 26, 2013 at 19:40 | |||||
| Sep 26, 2013 at 19:34 | history | edited | gung - Reinstate Monica | CC BY-SA 3.0 | edited for English; light formatting |
| Sep 26, 2013 at 19:27 | comment | added | gung - Reinstate Monica | The logit / logistic function is not the only function that can be used as a link function for regression models when the response is distributed as a binomial. Regarding this point, it may help you to read my answer here: difference-between-logit-and-probit-models. | |
| Sep 26, 2013 at 19:15 | history | asked | user16168 | CC BY-SA 3.0 |