Is there any role of redundancy analysis (for example using the redun() function of the Hmisc package in R) in finding variables to be included for a regression model? It seems logical but are there any points to be kept in mind?
Asked
Active
Viewed 267 times
2
Scortchi - Reinstate Monica
- 27,560
- 8
- 81
- 248
rnso
- 8,893
- 14
- 50
- 94
1 Answers
2
It's extremely useful, especially as it allows for non-monotonic relationships between predictors, but the caveats given in Low variance components in PCA, are they really just noise? Is there any way to test for it? & Examples of PCA where PCs with low variance are “useful” apply—the assumption that if two candidate predictors contain much the same information the difference between them is unimportant for explaining the response is one that has to be judged using subject-matter knowledge.
Scortchi - Reinstate Monica
- 27,560
- 8
- 81
- 248
-
I believe only predictor variables need be put in redundancy analysis. The dependent variable should not be put in the list for this analysis. Is that correct? – rnso Jun 03 '15 at 17:07
-
2Yes, that's correct. – Scortchi - Reinstate Monica Jun 03 '15 at 17:52