GetVariableGroupImpact: Model Implied Group Impact Analysis

In google/glmmplus: Multiple Imputation and Model Selection

Description

Shows how the predicted values of a model change as a set of variables simultaneously change in value.

Usage

GetVariableGroupImpact(gfo.obj, var.vec, low.val.list, high.val.list)

Arguments

gfo.obj	A gfo object created by FitModel, BackwardEliminate, or Forward Select
var.vec	A character vector of variable names
low.val.list	A list mapping each variable name (from var.vec) to its "low" value
high.val.list	A list mapping each variable name (from var.vec) to its "high" value

Details

The procedure works like this...

Examples

data(testdata)

# A sample data set with testdata values
head(testdata)

# Backwards elimination for fixed effect models 
gfo <- FitModel(y ~ x + w + z, data = testdata)
GetVariableGroupImpact(gfo, var.vec = c("w", "x"),
		       low.val.list = list(x = -1, w = -1),
		       high.val.list = list(x = 1, w = 1))

google/glmmplus documentation built on May 17, 2019, 7:47 a.m.