logisticDifferenceDf: Logistic Difference for Data Frames

In mdavis29/modelToolKit: Predictive Models for basic tasks at MUSC

Description

a function that interates through model predictors to estimate impact

Usage

logisticDifferenceDf(model, testObs, refData, className = NULL, p = 0.1,
  seed = 2012, verbose = TRUE)

Arguments

model	a model of class glm, rpart or ranger
testObs	a data frame with one row to predict with the model
refData	a data frame with columns included in model
className	name of the class column used for stratitifed resampling
p	percent of the refData to keep in the sample
seed	random sampleing seed
verbose	print debugging output

Value

a vector of estimated absolute influences

Author(s)

Matthew Davis

mdavis29/modelToolKit documentation built on May 22, 2019, 3:22 p.m.