roctab.hlp

是一个经济学管理应用软件 很难找的 但是经济学学生又必须用到

{smcl}
{* 03may2005}{...}
{cmd:help roctab}, {cmd:help roccomp}, {cmd:help rocgold}{...}
{right:dialogs:  {bf:{dialog roctab}}  {bf:{dialog roccomp}}}
{right:{bf:{dialog rocgold}} {space 7}}
{hline}

{title:Title}

{p2colset 5 16 18 2}{...}
{p2col:{hi:[R] roc} {hline 2}}Receiver-Operating-Characteristic (ROC) analysis
{p_end}
{p2colreset}{...}


{title:Syntax}

{pstd}
Perform nonparametric ROC analysis

{p 8 16 2}
{cmd:roctab}{space 1}
{it:refvar}
{it:classvar}
{ifin}
{weight}
[{cmd:,} {it:{help roctab##roctab_options:roctab_options}}]

{pstd}
Test equality of ROC areas

{p 8 16 2}
{cmd:roccomp}
{it:refvar}
{it:classvar}
[{it:classvars}]
{ifin}
{weight}
[{cmd:,} {it:{help roctab##roccomp_options:roccomp_options}}]

{pstd}
Test equality of ROC area against a standard ROC curve

{p 8 16 2}
{cmd:rocgold}
{it:refvar}
{it:goldvar}
{it:classvar}
[{it:classvars}]
{ifin}
{weight}
[{cmd:,} {it:{help roctab##rocgold_options:rocgold_options}}]


{marker roctab_options}{...}
{synoptset 25 tabbed}{...}
{synopthdr:roctab_options}
{synoptline}
{syntab:Main}
{synopt:{opt lor:enz}}report Gini and Pietra indices{p_end}
{synopt:{opt bam:ber}}calculate standard errors using the Bamber
method{p_end}
{synopt:{opt han:ley}}calculate standard errors using the Hanley
method{p_end}
{synopt:{opt bino:mial}}calculate exact binomial confidence intervals{p_end}
{synopt:{opt tab:le}}display the raw data in a 2 x k contingency table{p_end}
{synopt:{opt d:etail}}show details on sensitivity/specificity for each
cutpoint{p_end}
{synopt:{opt l:evel(#)}}set confidence level; default is
{cmd:level(95)}{p_end}
{synopt:{opt g:raph}}graph the ROC curve{p_end}
{synopt:{opt noref:line}}suppress plotting of the 45-degree reference
line{p_end}
{synopt:{opt spec:ificity}}graph sensitivity versus specificity{p_end}
{synopt:{opt sum:mary}}report the area under the ROC curve{p_end}

{syntab:Plot}
{synopt:{cmdab:plotop:ts(}{it:{help roctab##plot_options:plot_options}}{cmd:)}}affect 
	rendition of the ROC curve{p_end}

{syntab:Reference line}
{synopt:{opth rlop:ts(cline_options)}}affect rendition of the reference
line{p_end}

{syntab:Add plot}
{synopt:{opth "addplot(addplot_option:plot)"}}add other plots to generated
graph{p_end}

{syntab:Y-Axis, X-Axis, Title, Caption, Legend, Overall}
{synopt:{it:{help twoway_options}}}any options other than {opt by()} documented
in {bind:{bf:[G]} {it:twoway_options}}{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}
{opt fweight}s are allowed; see {help weight}.

{marker roccomp_options}{...}
{synoptset 25 tabbed}{...}
{synopthdr:roccomp_options}
{synoptline}
{syntab:Main}
{synopt:{opth by(varname)}}split into groups by variable{p_end}
{synopt:{opt bin:ormal}}estimate areas using binormal distribution
assumption{p_end}
{synopt:{opt test(matname)}}use contrast matrix for comparing ROC
areas{p_end}
{synopt:{opt l:evel(#)}}set confidence level; default is
{cmd:level(95)}{p_end}
{synopt:{opt g:raph}}graph the ROC curve{p_end}
{synopt:{opt noref:line}}suppress plotting of the 45-degree reference
line{p_end}
{synopt:{opt sep:arate}}place each ROC curve on its own graph{p_end}
{synopt:{opt sum:mary}}report the area under the ROC curve{p_end}

{syntab:Plot 1, Plot 2, Plot 3}
{synopt:{cmdab:plot#:opts(}{it:{help roctab##plot_options:plot_options}}{cmd:)}}affect 
	rendition of the {it:#}th  ROC curve{p_end}

{syntab:Reference line}
{synopt:{opth rlop:ts(cline_options)}}affect rendition of the reference
line{p_end}

{syntab:Y-Axis, X-Axis, Title, Caption, Legend, Overall}
{synopt:{it:{help twoway_options}}}any options other than {opt by} documented
in {bind:{bf:[G]} {it:twoway_options}}{p_end}

{p2coldent:+ {cmdab:line:}{ul:{it:#}}{cmd:opts(}{it:{help cline_options}}{cmd:)}}affect rendition of the {it:#}th binormal fit line{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}+ {cmd:line}{it:#}{cmd:opts(}{it:cline_options}{cmd:)} does not
appear in the dialog box.{p_end}
{p 4 6 2}{opt fweight}s are allowed; see {help weight}.

{marker rocgold_options}{...}
{synoptset 25 tabbed}{...}
{synopthdr:rocgold_options}
{synoptline}
{syntab:Main}
{synopt:{opt sid:ak}}adjust the significance probability using Sidak's
method{p_end}
{synopt:{opt bin:ormal}}estimate areas using binormal distribution
assumption{p_end}
{synopt:{opt test(matname)}}use contrast matrix for comparing ROC
areas{p_end}
{synopt:{opt l:evel(#)}}set confidence level; default is
{cmd:level(95)}{p_end}
{synopt:{opt g:raph}}graph the ROC curve{p_end}
{synopt:{opt noref:line}}suppress plotting of the 45-degree reference
line{p_end}
{synopt:{opt sep:arate}}place each ROC curve on its own graph{p_end}
{synopt:{opt sum:mary}}report the area under the ROC curve{p_end}

{syntab:Plot 1, Plot 2, Plot 3}
{synopt:{cmdab:plot#:opts(}{it:{help roctab##plot_options:plot_options}}{cmd:)}}affect 
        rendition of the {it:#}th ROC curve; plot 1 is the gold standard{p_end}

{syntab:Reference line}
{synopt:{opth rlop:ts(cline_options)}}affect rendition of the reference
line{p_end}

{syntab:Y-Axis, X-Axis, Title, Caption, Legend, Overall}
{synopt:{it:{help twoway_options}}}any options other than {opt by} documented
in {bind:{bf:[G]} {it:twoway_options}}{p_end}

{p2coldent:+ {cmdab:line:}{ul:{it:#}}{cmd:opts(}{it:{help cline_options}}{cmd:)}}affect rendition of the {it:#}th binormal fit line{p_end}
{synoptline}
{p2colreset}{...}
{p 4 6 2}+ {cmd:line}{it:#}{cmd:opts(}{it:cline_options}{cmd:)} does not
appear in the dialog box.{p_end}
{p 4 6 2}{opt fweight}s are allowed; see {help weight}.


{marker plot_options}{...}
{synoptset 25}{...}
{synopthdr:plot_options}
{synoptline}
INCLUDE help gr_markopt2
INCLUDE help gr_clopt
{synoptline}


{title:Description}

{pstd}
The above commands are used to perform Receiver Operating Characteristic
(ROC) analyses with rating and discrete classification data.

{pstd}
The two variables {it:refvar} and {it:classvar} must be numeric. The
reference variable indicates the true state of the observation, such as
diseased and nondiseased or normal and abnormal, and must be coded 0 and 1.
The rating or outcome of the diagnostic test or test modality is recorded in
{it:classvar}, which must be at least ordinal, with higher values indicating
higher risk.

{pstd}
{opt roctab} performs nonparametric ROC analyses.  By default,
{opt roctab} calculates the area under the ROC curve.  Optionally,
{opt roctab} can plot the ROC curve, display the data in tabular form, and
produce Lorenz-like plots.

{pstd}
{opt roccomp} tests the equality of two or more ROC areas obtained from
applying two or more test modalities to the same sample or to independent
samples.  {opt roccomp} expects the data to be in wide form when comparing
areas estimated from the same sample and in long form for areas estimated
from independent samples.

{pstd}
{opt rocgold} independently tests the equality of the ROC area of each of
several test modalities, specified by {it:casevar}, against a "gold" standard
ROC curve, {it:goldvar}.  For each comparison, {opt rocgold} reports
the raw and the Bonferroni adjusted significance probability.
Optionally, Sidak's adjustment for multiple comparisons can be obtained.

{pstd}
See {helpb rocfit} for a command that fits maximum-likelihood ROC models.


{title:Options for roctab}

{dlgtab:Main}

{phang}
{opt lorenz} specifies that the Gini and Pietra indices be reported.
Optionally, {opt graph} will plot the Lorenz-like curve.

{phang}
{opt bamber} specifies that the standard error for the area under the
ROC curve be calculated using the method suggested by Bamber.
Otherwise, standard errors are obtained as suggested by DeLong, DeLong, and
Clarke-Pearson.

{phang}
{opt hanley} specifies that the standard error for the area under the
ROC curve be calculated using the method suggested by Hanley and McNeil.
Otherwise, standard errors are obtained as suggested by DeLong, DeLong, and
Clarke-Pearson.

{phang}
{opt binomial} specifies that exact binomial confidence intervals be
calculated.

{phang}
{opt table} outputs a 2 x k contingency table displaying the raw data.

{phang}
{opt level(#)} specifies the confidence level, as a percentage,
for the confidence intervals. The default is {cmd:level(95)} or as set by
{helpb set level}.

{phang}
{opt detail} outputs a table displaying the sensitivity, specificity, the
percent of subjects correctly classified, and two likelihood ratios for each
possible cutpoint of {it:classvar}.

{phang}
{opt graph} produces graphical output of the ROC curve. If {opt lorenz}
is specified, the graphic output of a Lorenz-like curve will be produced.

{phang}
{opt norefline} suppresses the plotting of the 45-degree reference line
from the graphical output of the ROC curve.

{phang}
{opt specificity} produces a graph of sensitivity versus specificity,
instead of sensitivity versus (1 - specificity).  {opt specificity} implies
{opt graph}.

{phang}
{opt summary} reports the area under the ROC curve, its standard error,
and its confidence interval. If {opt lorenz} is specified, Lorenz indices are
reported.  This option is only needed when also specifying {opt graph}.

{dlgtab:Plot}

{phang}
{cmd:plotopts(}{it:{help roctab##plot_options:plot_options}}{cmd:)}
affect the rendition of the ROC curve{hline 2}the plotted points connected
by lines.  The {it:plot_options} can affect the size and color of markers,
whether and how the markers are labeled, and whether and how the points are
connected; see {it:{help marker_options}}, {it:{help marker_label_options}},
and {it:{help cline_options}}.

{dlgtab:Reference line}

{phang}
{opth rlopts(cline_options)} affect the rendition of the reference line; see
{it:{help cline_options}}.

{dlgtab:Add plot}

{phang}
{opth addplot(plot)} provides a way to add other plots to the
generated graph; see {it:{help addplot_option}}.

{dlgtab:Y-Axis, X-Axis, Title, Caption, Legend, Overall}

{phang}
{it:twoway_options} are any of the options documented in
{it:{help twoway_options}}, excluding {opt by()}.  These include options for
titling the graph (see {it:{help title_options}}) and options for saving the
graph to disk (see {it:{help saving_option}}).


{title:Options for roccomp and rocgold}

{dlgtab:Main}

{phang}
{opth by(varname)} ({opt roccomp} only) is required when comparing independent
ROC areas.  The {opt by()} variable identifies the groups to be compared.

{phang}
{opt sidak} ({opt rocgold} only) requests that the significance probability be
adjusted for the effect of multiple comparisons using Sidak's method.
Bonferroni's adjustment is reported by default.

{phang}
{opt binormal} specifies that the areas under the ROC curves to be
compared should be estimated using the binormal distribution assumption. By
default, areas to be compared are computed using the trapezoidal rule.

{phang}
{opt test(matname)} specifies the contrast matrix to be used when comparing
ROC areas.  By default, the null hypothesis that all areas are equal is
tested.

{phang}
{opt level(#)} specifies the confidence level, as a percentage, for the
confidence intervals.  The default is {cmd:level(95)} or as set by
{helpb set level}.

{phang}
{opt graph} produces graphical output of the ROC curve.

{phang}
{opt norefline} suppresses plotting of the 45-degree reference line from the
graphical output of the ROC curve.

{phang}
{opt separate} is meaningful only with {opt roccomp} and specifies that each
ROC curve be placed on its own graph rather than one curve on top of the
other.

{phang}
{opt summary} reports the area under the ROC curve, its standard error, and
its confidence interval.  This option is only needed when also specifying
{opt graph}.

{dlgtab:Plot 1, Plot 2, Plot 3}

{phang}
{cmd:plot}{it:#}{cmd:opts(}{it:{help roctab##plot_options:plot_options}}{cmd:)} 
affect the rendition of the {it:#}th ROC curve{hline 2}the curve's plotted
points connected by lines.  The {it:plot_options} can affect the size and
color of markers, whether and how the markers are labeled, and whether and how
the points are connected; see {it:{help marker_options}}, 
{it:{help marker_label_options}}, and {it:{help cline_options}}.

{pmore}
For {cmd:rocgold}, {cmd:plot1opts()} are applied to the ROC for the gold
standard.

{dlgtab:Reference line}

{phang}
{opth rlopts(cline_options)} affect the rendition of the reference line; see
{it:{help cline_options}}.

{dlgtab:Y-Axis, X-Axis, Title, Caption, Legend, Overall}

{phang}
{it:twoway_options} are any of the options documented in 
{it:{help twoway_options}}.  These include options for titling the graph
(see {it:{help title_options}}), options for saving the graph to disk
(see {it:{help saving_option}}), and the {opt by()} option (see 
{it:{help by_option}}).


{phang}
The following options are available with {cmd:roccomp} and {cmd:rocgold} but
are not shown in the dialog boxes:

{phang}
{opth line#:opts(cline_options)} affect rendition of line representing the
        {it:#}th ROC curve drawn using the binormal distribution assumption;
        see {it:{help cline_options}}.  These lines are drawn only when
        the {cmd:binormal} option is specified.{p_end}


{title:Examples}

{phang}{cmd:. roctab disease rating}{p_end}
{phang}{cmd:. roctab disease rating, graph}{p_end}
{phang}{cmd:. roctab disease rating, graph summary}{p_end}
{phang}{cmd:. roctab disease rating [fw=pop]}{p_end}
{phang}{cmd:. roctab disease rating, table detail}{p_end}
{phang}{cmd:. roctab disease rating, lorenz}{p_end}
{phang}{cmd:. roctab disease rating, lorenz graph}{p_end}
{phang}{cmd:. roccomp disease rating, by (mod)}{p_end}
{phang}{cmd:. roccomp disease mod1 mod2 mod3 [fw=pop], test(comp)}{p_end}
{phang}{cmd:. roccomp disease mod1 mod2 [fw=pop], binormal}{p_end}
{phang}{cmd:. rocgold disease mod1 mod2 mod3 [fw=pop]}{p_end}
{phang}{cmd:. rocgold disease mod1 mod2 mod3 [fw=pop], sidak}


{title:Also see}

{psee}
Manual:  {bf:[R] roc}

{psee}
Online:  {helpb logistic}, {helpb rocfit}
{p_end}