GetSingleInputApcs: GetSingleInputApcs

In dchudz/predcomps: Average Predictive Comparisons

Description

makes predictive comparison summaries (both per unit input and impact, both absolute and signed) by forming an data frame of pairs with appropriate weights and then calling 'ComputeApcFromPairs'. Only works fore continuous inputs right now

Usage

GetSingleInputApcs(predictionFunction, X, u, v, ...)

Arguments

predictionFunction	this could be a function (which takes data frame and makes returns a vector of predictions) or an object of class 'lm', 'glm', or 'randomForest'
X	a data frame with all inputs
u	a string naming the input of interest
v	a string naming the other inputs
...	other arguments to be passed to 'GetPairs'

Value

a list with: signed (the usual Apc) and absolute (Apc applied to the absolute value of the differences)

Examples

n <- 200
x1 <- runif(n = n, min = 0, max = 1)
x2 <- runif(n = n, min = 0, max = 1)
x3 <- runif(n = n, min = 0, max = 10)
y <- 2 * x1 + (-2) * x2 + 1 * x3 + rnorm(n, sd = 0.1)
df <- data.frame(x1, x2, x3, y)
fittedLm <- lm(y ~ ., data = df)
fittedLm
GetSingleInputApcs(fittedLm, df, "x2", c("x1", "x3"))

dchudz/predcomps documentation built on May 15, 2019, 1:48 a.m.