Lecture9 - Lecture 9. Modern Regression Suppose that we...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Lecture 9. Modern Regression Suppose that we consider a multiple linear regression model Y = β 0 + p X j =1 β j X j + ε, j = 1 , 2 ,...,p (8.1) The relationship between Y and regressors X j is frequently to be described by non-trivial non-linear functions. The Alternative Conditional Expectation (ACE) algorithm smoothly transforms X j in order to maximize the correlation between the response and the transformed predictors. In particular, the model defined by (8.1) may be replaced by the linear additive model Y = α + p X j =1 f j ( X j ) + ε, j = 1 , 2 , 3 . (8.2) Here f j are some smooth non-linear functions. Since f j is rather general, the new model 8.2 obviously subsumes the old model 8.1. The overall goal of ACE is to ”linearize” the relationship between the predictand Y and predictors X j . In particular, ACE finds transformations of predictors, f j , that make the re- lationship between Y and ”new” transformed predictors f j ( X j ) as linear as possible and automatically selects such smooth functions
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 01/12/2012 for the course STAT 331 taught by Professor Yuliagel during the Spring '08 term at Waterloo.

Page1 / 3

Lecture9 - Lecture 9. Modern Regression Suppose that we...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online