Conditional Distance Correlation Based Feature Screening and Conditional Independence Inference

Conditional distance correlation is a novel conditional dependence measurement of two multivariate random variables given a confounding variable. This package provides conditional distance correlation, performs the conditional distance correlation sure independence screening procedure for ultrahigh dimensional data < https://www3.stat.sinica.edu.tw/statistica/J28N1/J28N114/J28N114.html>, and conducts conditional distance covariance test for conditional independence assumption of two multivariate variable.

CDC Statistics

Jin Zhu

Installation

You can install the released version of cdcsis from CRAN with:

install.packages("cdcsis")

Example

Feature screening

This is a basic example which shows you how to pick out the important feature from high-dimensional dataset:

library(cdcsis)
set.seed(1)
num <- 100
p <- 1000
x <- matrix(rnorm(num * p), nrow = num, ncol = p)
z <- rnorm(num)
y <- 3*x[, 1] + 1.5*x[, 2] + 4*z*x[, 5] + rnorm(num)
res <- cdcsis(x, y, z)
head(res[["ix"]], n = 10)

cdcsis function successfully selects the informative variables from 1000 features pool.

[1]   1   5   2 628  17  87 912 903 395 630

License

GPL (>= 2)