Model-Free Knockoff Filter for Controlled Variable Selection

Model-free knockoffs are a general procedure for controlling the false discovery rate (FDR) when performing variable selection. For more information, see the website below and the accompanying paper: Candes et al., "Panning for Gold: Model-free Knockoffs for High-dimensional Controlled Variable Selection", 2016, .


MFKnockoff 0.9 (07/07/2017)

Initial release

MFKnockoff 0.9.1 (09/27/2017)

Improved stability and performance of SDP knockoffs, switching SDP solver from SCS to RDSDP Improved stability and performance of positive-definite checks Using approximate-SDP knockoffs by default (when p>500), instead of SDP

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.9.1 by Matteo Sesia, 2 years ago

Browse source code at

Authors: Rina Foygel Barber [ctb] (Development of the original fixed-design Knockoffs) , Emmanuel Candes [ctb] (Development of Model-Free Knockoffs and original fixed-design Knockoffs) , Lucas Janson [ctb] (Development of Model-Free Knockoffs) , Evan Patterson [aut] (Original R package for the original fixed-design Knockoffs) , Matteo Sesia [aut, cre] (R package for Model-Free Knockoffs)

Documentation:   PDF Manual  

GPL-3 license

Imports Rdsdp, Matrix, corpcor, glmnet, RSpectra, gtools

Depends on methods, stats

Suggests knitr, testthat, rmarkdown, lars, ranger, stabs, flare, doMC, parallel

See at CRAN