sens.cvasd: Get subgroup of sensitive patiens according to the "cvasd"...

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

Get subgroup of sensitive patiens according to the "cvasd" method.

Description

For a vector of binary responses, get sensitive patients by CVASD according to the input tuning set (eta.in, R.in, G.in). In the training subset, fit logistic regression for each covariate. In the testing subset, compute the new vs control arm odds ratio for covariates that have treatment-covariate interaction significant at a level eta.in in the training subset. A patient is sensitive if the odds ration > R.in for at least G.in covariates. If the length of each parameter in the tuning set (eta.in, R.in and G.in) > 1 then the tuning set is found by a nested cross-validation where only one inner fold is used (get.tuning.param function).

Usage

sens.cvasd(patients, covar, y, eta.in, R.in, G.in, seed)

Arguments

patients

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

Value

A list of 3 : psens - sensitivity of identifying the sensitive group, one value per simulation run pspec - specificity of identifying the sensitive group, one value per simulation run sens.pred - predicted sensitivity status (rows = patienst, columns = simulations)

Author(s)

Svetlana Cherlin, James Wason

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