plot.FPE: AIC-BIC instability plot

In tanujitdey/modelSampler: A unique tool for variable selection and model exploration in linear regression model

Description

Plot to visualize the instability of FPE models.

Usage

    plot.FPE(x, ...)

Arguments

x	An object of class boot.modelSampler, as that created by the function boot.modelSampler. See below.
...	Further arguments passed to or from other methods.

Details

When modelSampler is being called in the boot.modelSampler, each FPE model is kept track of in order to study the instability of these procedures. After finishing total bootstrap iteration, all FPE models are compared to the FPE models produced from full data set. This total comparison is visualized through a simple bar plot. Positive values less than 1 indicate that a given variable was selected by the FPE criteria in both the full data set and bootstrapped data set, but was not selected at all times during bootstrap iterations. Negative values greater than -1 imply that a particular variable is selected several times by the FPE criteria for bootstrapped data but was not selected in the full data set.

Author(s)

Tanujit Dey [e-mail]

References

Dey, T. (2013). modelSampler: An R Tool for Variable Selection and Model Exploration in Linear Regression. Journal of Data Science, 11(2), 371-387.

See Also

boot.modelSampler, modelSampler, print.boot.modelSampler, print.modelSampler, plot.ooberror, plot.icicle, plot.var.stability, plot.modelSampler.

Examples

  data(Pollute, package = "modelSampler") 
  ms.boot <- boot.modelSampler(MortRate~., Pollute, n.iter1=2500, 
  n.iter2=2500, B=20, verbose = TRUE)
  plot.FPE(ms.boot)

tanujitdey/modelSampler documentation built on May 5, 2019, 11:01 p.m.