plot.var.stability: Variable stability plot

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

Description

Variable stability plot over the model space.

Usage

    plot.var.stability(
		x, 
		filter.flag = FALSE,
		high.dim = 25,
		...)

Arguments

x	An object of class boot.modelSampler, created by the function boot.modelSampler. See below.
filter.flag	It is an optional argument. Default value is FALSE. If the value is TRUE, then it plots only the best subset of variables over the entire model space. It's useful in high dimensional data. When number of variables are very high, it's hard to visualize the path important variables over the model space. This argument will help user to see the path of the best subset of variables.
high.dim	An optional argument. 25 is default setting. Usful in high dimensional setting. This argument implements median smoothing to the estimators to enchance better graphical depiction.
...	Further arguments passed to or from other methods.

Details

Coming soon...

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.modelSampler, plot.icicle, plot.ooberror, plot.FPE.

Examples

#---------------------------------------------------------------------------
# Example 1:
# In the following example Ozone data set is used to draw the plot. Total number
# of variables in the data set is 12. Therefore optional arguments are not so
# useful to use.
#----------------------------------------------------------------------------

data(Ozone, package = "modelSampler") 
ms.boot <- boot.modelSampler(ozone~., 
			 Ozone, 
			 n.iter1=2500, 
  			 n.iter2=2500, 
			 B=20, 
			 verbose = TRUE)

plot.var.stability(ms.boot, 
			filter.flag=FALSE, 
			high.dim = 25)

#--------------------------------------------------------------------------------------
# Example 2:
# In this example Ozone interaction data set is used. Total number of variables in this
# data set is 89. So, it's now useful to use optional arguments. This example will plot 
# only the best subset of variables over the modified model space.
#---------------------------------------------------------------------------------------

data(OzoneI, package = "modelSampler") 
ms.boot <- boot.modelSampler(ozone~., 
			 OzoneI, 
			 n.iter1=2500, 
  			 n.iter2=2500, 
			 B=20, 
			 verbose = TRUE)

plot.var.stability(ms.boot, 
			filter.flag=TRUE, 
			high.dim = 25)

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