sig_precision_crude: Crude estimate of precision gain due to gene signature...

In leekgroup/sig2trial: Pipeline R package for creating simple gene signatures and testing their value in a clinical trial setting

Description

This function takes training/test data and pairs generated via empirical control feature selection and builds a decision tree model. It also cross-validates to get an out-of-sample accuracy estimate

Usage

sig_precision_crude(test_outcome, test_pred, test_covar)

Arguments

test_outcome	A vector of length n outcomes for the test/validation set. Currently only supports binary outcomes.
test_pred	Prediction vector of length n created using the decision tree model built by tsp_model_builder.R
test_covar	A n x p matrix of additional covariates to adjust for. (optional)

Details

This function approximates how much precision gain we might expect in a clinical trial setting if we adjused for predictions with our gene signature. We can optionally provide additional covariates and estimate how much additional gain we might observe if we include predictions from our model.

Value

A crude approximation of the reduction in variance of a treatment effect estimator for the outcome of interest.

leekgroup/sig2trial documentation built on May 20, 2019, 11:31 p.m.