Timeline for "matrix is not positive definite" - even when highly correlated variables are removed
Current License: CC BY-SA 3.0
2 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Mar 6, 2013 at 16:27 | comment | added | meepmeep | I am working with 14 variables, with n = 10 for each correlation. Shouldn't I be able to tell which variables are redundant by looking at the correlation coefficients? And if so, is it possible for two vbls to be redundant with r = 0.8 or less? I thought r would have to be high (>0.9) for them to be redundant. | |
| Mar 6, 2013 at 15:32 | history | answered | hydrologist | CC BY-SA 3.0 |