Plsdaroc

Purpose

Calculate and display ROC curves for a PLSDA model.

Synopsis

roc = plsdaroc(model,ycol,options)

Description

ROC curves can be used to visualize the specificities and sensitivities that are possible with different predicted y-value thresholds in a PLSDA model.

Inputs

• model = a PLSDA model structure

Optional Inputs

• ycol = an optional index into the y-columns used in the PLSDA model ycol [default = all columns],
• options = options structure (see below)

Outputs

• roc = dataset with the sensitivity/specificity results that are needed to plot ROC curves.

Options

options = options structure that can contain the following fields

• plots : [ 'none' | {'final'}] governs plotting on/off
• figure : [ 'new' | 'gui' | figure_handle ] governs location for plot
  • 'new' plots onto a new figure
  • 'gui' plots using noninteger figure handle
  • A figure handle 'figure_handle' specifies the figure onto which the plot should be made.
• plotstyle : [ 'roc' | 'threshold' | {'all'} ] governs type of plots.
  • 'roc' and 'threshold' give only the specified type of plot
  • 'all' shows both types of plots on one figure (default).
  • plotstyle can also be specified as '1' (which gives 'roc' plots) or 2 (which gives 'threshold' plots).
• showauc : [ {'on'} | 'off'] controls drawing AUC value on ROC plot. Note, clicking on the AUC text in the plot will remove it.

See Also

discrimprob, plsda, plsdthres, simca, roccurve