getThresholds: Get prediction risk thresholds for quantiles

Description Usage Arguments Details Value

View source: R/metrics.R

Description

This function calculates the net reclassification improvement (NRI) metric comparing two models at a range of predicting tresholds

Usage

1
getThresholds(plpModel1, plpModel2, percentage = c(0.99, 0.95, 0.9, 0.5))

Arguments

plpModel1

The object returned by runPlp() containing the trained model or the output when implementing an existing model

plpModel2

The object returned by runPlp() containing the trained model or the output when implementing an existing model

percentage

A quantile percentage (e.g., c(0.99, 0.95)) will find the threshold that only 1 percent and 5 percent of people have a risk equal to or higher than for both models

Details

Users need to input a trained model (the output of runPlp()) or the output of running an existing model

Value

A data.frame with each column being the risk thresholds per model and rows corresponding to the input quantiles


OHDSI/PredictionComparison documentation built on July 14, 2020, 6:16 p.m.