help metandiplot(Roger Harbord) also see: metandi -------------------------------------------------------------------------------

Title

metandiplot-- SROC plot of results from metandi

Syntax

metandiplot[tp fp fn tn] [if] [in] [weight] [,notruncatelevel(#)predlevel(numlist)npoints(#)subplot_optionsaddplot(plot)twoway_options]

aweights,fweights, andpweights are allowed; see weight.

Description

metandiplotgraphs the results frommetandion a summary receiver operating characteristic (SROC) plot. By default the display includes:- a

summary pointshowing the summary sensitivity and specificity - aconfidence contouroutlining the confidence region for the summary point - one or moreprediction contoursoutlining the prediction region for the true sensitivity and specificity in a future study - theHSROC curvefrom the hierarchical Summary ROC (HSROC) modelIf the optional variables

tp fp fn tnare included on the command line, the plot also includesstudy estimatesindicating the sensitivity and specificity estimated using the data from each study separately.Any of these features may be customised or turned off using the subplot_options.

Options

notruncatespecificies that the HSROC curve willnotbe truncated outside the region of the data. By default, the HSROC curve is not shown when the sensitivity or specificity is less than its smallest study estimate.

level(#)specifies the confidence level, in percent, for the confidence contour; see help level.

predlevel(numlist)specifies the levels, in percent, for the prediction contour(s). The default is a single contour at the same probability level as the confidence region. Up to five prediction contours are allowed.

npoints(#)specifies the number of points to use in drawing the outlines of the confidence and prediction regions. The default is 500.

subplot_options:summopts(),confopts(),predopts(),curveopts()andstudyopts()specify options that control the display of the summary point, confidence contour, prediction contour(s), HSROC curve and study symbols respectively. The options within each set of parantheses are simply passed through to the appropriatetwowayplot. In addition, any of the plots can be turned off by specifying, for example,summopts(off).

addplot(plot)allows adding additionalgraphtwowayplots to the graph; seeaddplot_option. For example, empirical Bayes predictions could be generated by usingpredictaftermetandiand added to the graph. See metandipostestimation.

twoway_optionsare most of the options documented intwoway_options, including options for titles, axes, labels, schemes and saving the graph to disk. Theby()option is not allowed, however.

RemarksThe default is to weight the study estimates by the total number in each study, giving symbols (open circles by default) scaled according to the size of the weights; see weighted markers. To make the symbols all the same size, specify constant weights, e.g.

[aw=1].

Examples

. metandiplot

. metandiplot tp fp fn tn

. metandiplot tp fp fn tn [aw=1], conf(off) curve(off) predlevel(50 80 9599)

Also see