get.tuning.param: Get tuning parameters for CVASD by nested cross-validation.

In svetlanache/rapids: What the Package Does (One Line, Title Case)

Get tuning parameters for CVASD by nested cross-validation.

Description

An internal function for tuning the parameters. In one inner fold only, compute p-value for the test of treatment effect comparison in the selected sensitive group, for each combination of (eta, R, G). Output the (eta, R, G) combination that corresponds to the smallest p-value.

Usage

get.tuning.param(patients.train, covar.train, y.train, eta, R, G, seed)

Arguments

patients

- a data frame of patients inormation covar - a data frame of covariates y - a vector of responses eta - significance level for covariate-wise logistic regression for "cvasd" method (double of vector of doubles) R - a threshold of the odds ratio for "cvasd" method (double of vector of doubles) G - a threshold for the number of covar for "cvasd" method (integer of vector of integers) seed - a seed for random number generator

Value

A data frame with the optimal eta, R, and G

Author(s)

Svetlana Cherlin, James Wason

svetlanache/rapids documentation built on Sept. 15, 2023, 7 a.m.