Post-selection inference for L1-penalized likelihood models

News

  • Author: Statistics Views, Jonathan Taylor & Robert Tibshirani
  • Date: 19 January 2018

A new article available from Early View in the Canadian Journal of Statistics presents present a new method for post-selection inference for math formula (lasso)'penalized likelihood models, including generalized regression models. The approach by Jonathan Taylor and Robert Tibshirani generalizes the post-selection framework presented in Lee et al. (2013) and the authors explain their findings below.

thumbnail image: Post-selection inference for L1-penalized likelihood models

We present a new method for post-selection inference for L1 (lasso)-penalized likelihood models, including generalized regression models and Cox's proportional hazards model for survival data . The method provides p-values and confidence intervals that properly account for the selection of predictors in the modelling process. This work enables data analysts to avoid spurious findings due to overfitting the data.

The proposals of this article are implemented in our selectiveInference R package in the public CRAN repository.

Post-selection inference for L1-penalized likelihood models 

Jonathan Taylor & Robert Tibshirani

Canadian Journal of Statistics, Early View

DOI: 10.1002/cjs.11313

Related Topics

Related Publications

Related Content

Site Footer

Address:

This website is provided by John Wiley & Sons Limited, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ (Company No: 00641132, VAT No: 376766987)

Published features on StatisticsViews.com are checked for statistical accuracy by a panel from the European Network for Business and Industrial Statistics (ENBIS)   to whom Wiley and StatisticsViews.com express their gratitude. This panel are: Ron Kenett, David Steinberg, Shirley Coleman, Irena Ograjenšek, Fabrizio Ruggeri, Rainer Göb, Philippe Castagliola, Xavier Tort-Martorell, Bart De Ketelaere, Antonio Pievatolo, Martina Vandebroek, Lance Mitchell, Gilbert Saporta, Helmut Waldl and Stelios Psarakis.