Timeline for How to best model interaction effect of two continuous predictor variables?
Current License: CC BY-SA 3.0
9 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Feb 1, 2016 at 22:44 | vote | accept | Clark Chong | ||
| S Feb 1, 2016 at 13:06 | history | suggested | Henry.L |
Add tags for further attention to this problem.
|
|
| Feb 1, 2016 at 12:43 | answer | added | Henry.L | timeline score: 1 | |
| Feb 1, 2016 at 12:11 | review | Suggested edits | |||
| S Feb 1, 2016 at 13:06 | |||||
| Feb 1, 2016 at 6:22 | comment | added | Clark Chong | @Henry.L: would you mind helping me understand what you mean by $C^2$ and risk function in your comment? A reference or even a worked example would be very helpful! Thanks! | |
| Jan 31, 2016 at 14:30 | comment | added | Henry.L | A possible way of doing this is by allowing more flexible forms of interaction terms. For example you might want the interaction be, say $\mathcal{C}^{2}$, and then add a penalized term in the risk function when you evaluate the regression model. This approach yields satisfying results and is still considered as a research area. | |
| Jan 31, 2016 at 13:24 | answer | added | Kim Castelin | timeline score: -1 | |
| Jan 31, 2016 at 13:21 | answer | added | Frank Harrell | timeline score: 5 | |
| Jan 31, 2016 at 7:55 | history | asked | Clark Chong | CC BY-SA 3.0 |