Here is a new instance (and an implementation) of LASSO: Statistical Estimation and Testing via the Ordered L1 Norm by Malgorzata Bogdan, Ewout van den Berg Weijie Su, Emmanuel J. Candes
The attendant code in Matlab and R is located here.We introduce a novel method for sparse regression and variable selection, which is inspired by modern ideas in multiple testing. Imagine we have observations from the linear model , then we suggest estimating the regression coefficients by means of a new estimator called the ordered lasso, which is the solution tohere, and is the order statistic of the magnitudes of . In short, the regularizer is an ordered norm which penalizes the regression coefficients according to their rank: the higher the rank — the closer to the top — the larger the penalty. This is similar to the famous Benjamini-Hochberg procedure (BHq) , which compares the value of a test statistic taken from a family to a critical threshold that depends on its rank in the family. The ordered lasso is a convex program and we demonstrate an efficient algorithm for computing the solution. We prove that for orthogonal designs with variables, taking ( is the cumulative distribution function of the errors), , controls the false discovery rate (FDR) for variable selection. This holds under the assumption that the errors are i.i.d. symmetric and continuous random variables. When the design matrix is nonorthogonal there are inherent limitations on the FDR level and the power which can be obtained with model selection methods based on -like penalties. However, whenever the columns of the design matrix are not strongly correlated, we demonstrate empirically that it is possible to select the parameters as to obtain FDR control at a reasonable level as long as the number of nonzero coefficients is not too large. At the same time, the procedure exhibits increased power over the lasso, which treats all coefficients equally. The paper illustrates further estimation properties of the new selection rule through comprehensive simulation studies.
Liked this entry ? subscribe to Nuit Blanche's feed, there's more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.