From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search


Calculate and display ROC curves for a PLSDA model.


roc = plsdaroc(model,ycol,options)


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


  • 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)


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


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