A sensitivity analysis and error bounds for the adaptive lasso.

Abstract

Sparse regression is an efficient statistical modelling technique which is of major relevance for high dimensional problems. There are several ways of achieving sparse regression, the well-known lasso being one of them. However, lasso variable selection may not be consistent in selecting the true sparse model. Zou (2006) proposed an adaptive form of the lasso which overcomes this issue, and showed that data driven weights on the penalty term will result in a consistent variable selection procedure. Weights can be informed by a prior execution of least squares or ridge regression. Using a power parameter on the weights, we carry out a sensitivity analysis for this parameter, and derive novel error bounds for the Adaptive lasso.