stat_glmnet: Importance statistics: Absolute elastic-net coefficient...

In kormama1/seqknockoff: Sequential knockoffs for continuous and categorical variables

Description

This is a wrapper function that calls knockoff::glmnet.stat_coefdiff. The input data.frames are first converted to design matrices (with the function model.matrix). This means that if the input features contain factor variables then an importance statistic is calculated for each dummy variable as determined by the model.matrix contrasts (defaults to indicator dummy variables with a reference level).

Usage

stat_glmnet(X, X_k, y, family, ...)

Arguments

X	original data.frame (or tibble) with "numeric" and "factor" columns only. The number of columns, ncol(X) needs to be > 2.
X_k	knockoff data.frame (or tibble) with "numeric" and "factor" columns only. The dimensions and column classes must match those of the original X.
y	response vector with length(y) = nrow(X). Accepts "numeric" (family="gaussian") or binary "factor" (family="binomial").
family	should be "gaussian" if y is numeric, but "binomial" if y is a binary factor variable.
...	additional parameters passed to knockoff::stat.glmnet_coefdiff

Value

a vector of importance statistics W of length equal to number of columns of the model.matrix of X.

Examples

library(seqknockoff)

set.seed(1)

# Simulate 10 Gaussian covariate predictors:
X <- generate_X(n=1000, p=10, p_b=0, cov_type="cov_equi", rho=0.5)

 # Calculate the knockoff copy of X:
 X_k <- knockoffs(X)

# Simulate response from model y = X%*%beta + epsilon, where epsilon ~ N(0,1) with
# first 5 beta-coefficients = 8 (all other zero).
y <- generate_y(X, p_nn=5, a=8)

W <- stat_glmnet(X=X, X_k=X_k, y=y, family="gaussian", nfolds=5)

kormama1/seqknockoff documentation built on April 11, 2021, 7:44 a.m.