cv_intscreen: Cross validation interaction screening

In jaredhuling/intscreen:

Description

This function implements CV interaction screening

Usage

cv_intscreen(x, y, nints = 100, nsplits = 10, train.frac = 0.75,
  fraction.in.thresh = 1, resample.type = c("bootstrap", "cv"),
  heredity = c("none", "weak", "strong"), verbose = FALSE,
  modifier = NULL)

Arguments

x	matrix of predictors
y	vector of observations
nints	integer number of top interactions to screen
nsplits	integer number of cross validation splits to run. defaults to 10
train.frac	fraction of data used for each split. defaults to 0.75. Only used for resample.type = "cv"
fraction.in.thresh	fraction of times across the nsplits CV splits each interaction is required in the top k interactions in order to be selected
resample.type	either "cv" for cross validation or "bootstrap" for bootstrap approach
heredity	either "weak", "strong", or "none"
verbose	logical value, whether to print progress of the CV splitting
modifier	effect modifier

Examples

set.seed(1)
x <- matrix(rnorm(150 * 500), ncol = 500)
y <- rnorm(150) + x[,1] * x[,2] * 0.5 - x[,3] * x[,4] * 0.5

## require that each interaction be in the top 200 ints 5% of the 10 splits
ints <- cv_intscreen(x, y, nints = 200, nsplits = 10, fraction.in.thresh = 0.75)

ints$int_idx

## require that each interaction be in the top 50 ints 50% of the 10 splits
ints <- cv_intscreen(x, y, nints = 50, nsplits = 10, fraction.in.thresh = 0.5)

ints$int_idx

jaredhuling/intscreen documentation built on May 19, 2019, 9:38 p.m.