phimr: Permutation-Based High-Dimensional Multiple Response Filter

In brianpatrickneal/AMKAT: AMKAT: An Adaptive Multivariate Kernel-based Association Test

Permutation-Based High-Dimensional Multiple Response Filter

Description

Selects columns of x related to one or more columns of y using a filter method in which permuted data is compared to original data using nonparametric tests of association between columns from x and columns from y.

Usage

phimr(y, x)

Arguments

y	a numeric matrix containing data on the dependent variables, with at least two observations indexed by row.
x	a numeric matrix with the same number of rows as y, containing data on the independent variables.

Details

A two-sided test of the hypothesis that Spearman's Rho equals 0 is performed for each pair of columns, one from x and one from y. A row-permuted copy of x is created, and the tests are repeated using the permuted copy of x. A column of x passes the filter (i.e., is deemed to be related to y) if the minimum of the P-values for that column from the original data is less than the minimum of the P-values from the permuted data.

If no columns of x pass the filter, then the column with the lowest minimum P-value is selected by default.

Value

An integer vector of variable length, containing the indices of the columns of x which passed the filter.

Author(s)

Brian Neal

References

Neal, Brian and He, Tao. “An adaptive multivariate kernel-based test for association with multiple quantitative traits in high-dimensional data.” Genetic Epidemiology (not yet submitted).

Examples

y <- matrix(rnorm(4 * 25), nrow = 25, ncol = 4)
x <- matrix(rnorm(50 * 25), nrow = 25, ncol = 50)
selected_x_columns <- phimr(y, x)

brianpatrickneal/AMKAT documentation built on June 15, 2022, 8:47 a.m.