This preview shows page 1. Sign up to view the full content.
Unformatted text preview: 9.5 If the alpha-to-enter value exceeds the alpha-to-remove value we may end up with a situation where some variables keep getting added into the model (because their p-value is less than the alpha-to-enter value) and are later removed (because their p-value, recalculated with respect to the rest of the variables currently in the model, exceeds the alpha-to-remove value)....
View Full Document
This note was uploaded on 11/28/2009 for the course STAT 4315 taught by Professor Heuter during the Fall '09 term at Columbia.
- Fall '09
- Linear Regression