Reference

Generate Simulation Data
generate_X()	Simulate Gaussian and binary covariate predictors
generate_y()	Simulate Gaussian response from a sparse regression model
generate_lp()	Generate linear predictor with first p_nn beta coefficients = a, all other = 0
generate_Weib()	Function that simulates response from Cox model with Weibull baseline hazard.
Controlled Feature Selection
knockoff.filter()	The Knockoff Filter
knockoff.select()	Select Variables based on knockoff statistics
knockoff.threshold()	Threshold for the knockoff filter
print(<knockoff.filter>)	Print results for the multiple knockoff filter
Generate Knockoffs
create.shrink_Gaussian()	Create Shrink Gaussian Knockoffs
create.sparse_Gaussian()	Create Sparse Gaussian Knockoffs
create.pc()	Create PC Knockoffs
create.pls()	Generate knockoff variables using PLS regression (PLSKO)
create.zpls()	Generate knockoff variables using sparse partial least squares (SPLS) regression
create.seq()	Sequential knockoffs for continuous and categorical variables
create.sparse_seq()	Sparse sequential knockoff generation algorithm
Calculate Knockoffs Statistics
stat.SHAP()	Importance statistics based on XGBoost
stat.forward_selection()	Importance statistics based on forward selection
stat.glmnet_coefdiff()	Importance statistics based on a GLM with cross-validation
stat.glmnet_lambdadiff()	Importance statistics based on a GLM
stat.glmnet_lambdasmax()	GLM statistics for knockoff
stat.random_forest()	Importance statistics based on random forests
stat.sqrt_lasso()	Importance statistics based on the square-root lasso
stat.stability_selection()	Importance statistics based on stability selection
stat.xgboost()	Importance statistics based on xgboost
Recovery functions
glmnet.recovery()	Simulate from glmnet penalized regression model
ols.recovery()	Calculate \(\hat{X}\) by fitting OLS regression on its neighbors
pls.recovery()	Calculate \(\hat{X}\) by fitting PLS regression on its neighbours
simple.recovery()	Simple knockoff generator
spls.recovery()	Calculate \(\hat{X}\) by fitting sparse PLS
Aggregate Multiple Knockoffs
agg_Avg()	Aggregated Knockoff with Average Test Statistics
agg_BH()	Aggregated Knockoffs with AKO (Aggregation of Multiple Knockoffs)
agg_Freq()	Aggregated Knockoff Using Selection Frequency
Performance Evaluation
perf_eval()	Evaluate False Discovery Proportion (FDP) and True Positive Proportion (TPP)