sing.mod: Single MDR Model

In CateSelection: Categorical Variable Selection Methods

Description

Computes coefficient of determination and p-value for each single (marginal) MDR model

Usage

sing.mod(x, y, order = NULL, alpha = NULL, beta = NULL, delete = NULL, trace = NULL, ...)

Arguments

x	A matrix of genotypic data/genetic markers (predictor variables), where the rows are the samples and the columns are the predictors.
y	A numeric vector of phenotypic data (response variable).
order	The order of interaction. Default is 2.
alpha	Cutoff value for p-value. Default is 0.01. If "delete" is TURE, the interactive predictors in the model with the p-value greater than "alpha" will be removed in the final results.
beta	Cutoff value for coefficient of determination. Default is 0.01. If "delete" is TURE, the interactive predictors in the model with the coefficient of determination smaller than "beta" will be removed in the final results.
delete	Logistic value. Default FALSE; TRUE will remove the interactive predictors according two cutoff values "alpha" and "beta".
trace	Show computations? Default FALSE.
...	Other arguments for future methods.

Value

It returns a matrix with the index of interactive predictors, and the corresponding p-value, coefficient of determination, and adjusted coefficient of determination.

References

Yi Xu, Jixiang Wu, Detecting epistatic effects among SNP markers associated with three barley traits by a MDR based forward selection method (unpublished).

Examples

data(data1)
y <- data1[,1]
x <- data1[,-1]
res <- sing.mod(x,y,order=2)
res

CateSelection documentation built on May 1, 2019, 10:11 p.m.