rxStepPlot: Step Plot

Article
07/12/2022

Description

Plot stepwise coefficients for rxLinMod, rxLogit and rxGlm objects.

Usage

  rxStepPlot(x, plotStepBreaks = TRUE, omitZeroCoefs = TRUE, eps = .Machine$double.eps, 
    main = deparse(substitute(x)), xlab = "Step", ylab = "Stepwise Coefficients", 
    type = "b", pch = "*",
      ...  )

Arguments

`x`

a stepwise regression object of class rxLinMod, rxLogit, or rxGlmwith a non-empty stepCoefs component, which can be generated by setting the component keepStepCoefs of the variableSelection argument to TRUE when the model is fitted.

`plotStepBreaks`

logical value. If TRUE, vertical lines are drawn at each step in the stepwise coefficient paths.

`omitZeroCoefs`

logical flag. If TRUE, coefficients with sum of absolute values less than eps will be omitted for plotting.

`eps`

the smallest positive floating-point number that is treated as nonzero.

`main, xlab, ylab, type, pch, ...`

additional graphical arguments to be passed directly to the underlying matplot function.

Value

the nonzero stepwise coefficients are returned invisibly.

Author(s)

Microsoft Corporation Microsoft Technical Support

Examples


 ## setup
 form <- Sepal.Length ~ Sepal.Width + Petal.Length
 scope <- list(
     lower = ~ Sepal.Width,
     upper = ~ Sepal.Width + Petal.Length + Petal.Width * Species)

 ## rxLinMod/variableSelection with keepStepCoefs = TRUE
 varsel <- rxStepControl(method = "stepwise", scope = scope, keepStepCoefs = TRUE)
 rxlm.step <- rxLinMod(form, data = iris, variableSelection = varsel,
     verbose = 1, dropMain = FALSE, coefLabelStyle = "R")
 rxStepPlot(rxlm.step)

Share via