GhostKnockoff: The Knockoff Inference Using Summary Statistics

Functions for multiple knockoff inference using summary statistics, e.g. Z-scores. The knockoff inference is a general procedure for controlling the false discovery rate (FDR) when performing variable selection. This package provides a procedure which performs knockoff inference without ever constructing individual knockoffs (GhostKnockoff). It additionally supports multiple knockoff inference for improved stability and reproducibility. Moreover, it supports meta-analysis of multiple overlapping studies.

Version: 0.1.0
Depends: R (≥ 3.5.0), Matrix, CVXR, Rdsdp, gtools, seqminer
Imports: RSpectra, corpcor
Published: 2021-11-23
Author: Zihuai He
Maintainer: Zihuai He <zihuai at>
License: GPL-3
NeedsCompilation: no
CRAN checks: GhostKnockoff results


Reference manual: GhostKnockoff.pdf


Package source: GhostKnockoff_0.1.0.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): GhostKnockoff_0.1.0.tgz, r-release (x86_64): GhostKnockoff_0.1.0.tgz, r-oldrel: GhostKnockoff_0.1.0.tgz


Please use the canonical form to link to this page.