Forward stagewise regression and the monotone lasso

Details Forward stagewise regression and the monotone lasso Forward stagewise regression and the monotone lasso

2007-05-02

arxiv.org/abs/0705.0269v1

Electronic Journal of Statistics 2007, Vol. 1, 1-29

Mathematics

Statistics Theory

Published at http://dx.doi.org/10.1214/07-EJS004 in the Electronic Journal of Statistics (http://www.i-journals.org/ejs/) by t

Scientific paper

10.1214/07-EJS004

We consider the least angle regression and forward stagewise algorithms for solving penalized least squares regression problems. In Efron, Hastie, Johnstone & Tibshirani (2004) it is proved that the least angle regression algorithm, with a small modification, solves the lasso regression problem. Here we give an analogous result for incremental forward stagewise regression, showing that it solves a version of the lasso problem that enforces monotonicity. One consequence of this is as follows: while lasso makes optimal progress in terms of reducing the residual sum-of-squares per unit increase in $L_1$-norm of the coefficient $\beta$, forward stage-wise is optimal per unit $L_1$ arc-length traveled along the coefficient path. We also study a condition under which the coefficient paths of the lasso are monotone, and hence the different algorithms coincide. Finally, we compare the lasso and forward stagewise procedures in a simulation study involving a large number of correlated predictors.

Affiliated with

Also associated with

No associations

Rating

Forward stagewise regression and the monotone lasso does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Forward stagewise regression and the monotone lasso, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Forward stagewise regression and the monotone lasso will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-674635