plot.grpregOverlap

The fitted <code>"grpregOverlap"</code> object.


  Where to put the legend?  If left unspecified, no legend is drawn.  
  See <code>legend</code> for details.

legend.loc

Controls alpha-blending.  Default is alpha=1.

alpha


  Should plot the paths of latent coefficients? Default is TRUE. Note $\hat{\gamma}$
  is used as the label of y axis to represent latent coefficient vector.
  

latent

Should horizontal axis be on the log scale?  Default is FALSE.

log.l


  If <code>TRUE</code>, plot the norm of each group, rather than the individual 
  coefficients. Note that the norm is of latent coefficients.
  

norm


  Other graphical parameters to <code>plot</code>, <code>matlines</code>, or <code>legend</code>.
  


Plot the paths of estimated coefficients, latent coefficients (default), or L2 
norms of the groups for a fitted object "grpregOverlap".


Graphics

Fit the regularization path of linear, logistic or Cox models with
overlapping grouped covariates based on the latent group lasso approach. Latent
group MCP/SCAD as well as bi-level selection methods, namely the group exponential
lasso and the composite MCP are also available. This package serves as an
extension of R package 'grpreg' (by Dr. Patrick Breheny <patrick-breheny@uiowa.edu>)
for grouped variable selection involving overlaps between groups.

plot.grpregOverlap: Plot object "grpregOverlap"

Description

Usage

Arguments

See Also

Examples