Timeline for How to deal with high correlation among predictors in multiple regression?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 26, 2015 at 23:43 | comment | added | Glen_b | Indeed, there are a number of pronouncements in Tabachnick and Fidell that I'd regard as at least somewhat dubious ... just because something is printed in a book doesn't mean it always makes sense. | |
| Feb 25, 2015 at 20:33 | history | protected | whuber♦ | ||
| Oct 20, 2013 at 0:52 | history | edited | Scortchi♦ | CC BY-SA 3.0 |
fixed typos
|
| Sep 27, 2012 at 20:01 | history | edited | whuber♦ | CC BY-SA 3.0 |
appended answer 38137 as supplemental
|
| Sep 27, 2012 at 16:20 | comment | added | whuber♦ | For an explicit example of this situation, see the analysis of 10 IVs at stats.stackexchange.com/a/14528. Here, all the IVs are strongly correlated (around 60%). But if you excluded all of them, you wouldn't have anything left! Often it's the case that you cannot drop any of these variables. This makes the T&F recommendation untenable. | |
| Sep 27, 2012 at 11:56 | history | edited | chl | CC BY-SA 3.0 |
deleted 2 characters in body; edited title
|
| Sep 27, 2012 at 10:15 | answer | added | Peter Flom | timeline score: 30 | |
| Sep 27, 2012 at 9:17 | review | First posts | |||
| Sep 27, 2012 at 20:02 | |||||
| Sep 27, 2012 at 9:12 | history | asked | Ander | CC BY-SA 3.0 |