plot.proximity {randomSurvivalForest}R Documentation

Extract and Plot the Proximity Matrix

Description

Multidimensional scaling plot of the proximity matrix.

Usage

    plot.proximity(x, plot = TRUE, ...)

Arguments

x

An object of class (rsf, grow) or (rsf, predict).

plot

Should proximity be plotted.

...

Further arguments passed to or from other methods.

Details

Extracts the proximity information from the object x and transforms this to a symmetric proximity matrix. Dissimilarities between points are converted into distances using the multidimensional scaling function cmdscale and then plotted. Overlayed on the plot are mortality values, rescaled from 1-100, with 1 indicating low mortality, and 100 indicating high mortality. Mortality values will be well separated in successful analyses. Note that points in blue correspond to events, whereas black points are censored observations.

Value

Invisibly, the proximity matrix with entries transformed to relative frequencies.

Author(s)

Hemant Ishwaran hemant.ishwaran@gmail.com

Udaya B. Kogalur kogalurshear@gmail.com

See Also

rsf, predict.rsf.

Examples

  data(pbc, package = "randomSurvivalForest") 
  pbc.prox.out <- rsf(Surv(days,status) ~ ., pbc, ntree = 100, proximity = TRUE)
  plot.proximity(pbc.prox.out)

[Package randomSurvivalForest version 3.6.3 Index]