{smcl}
{* 08apr2010}{...}
{hline}
help for {hi:r2o}{right:(Michael Lacy)}
{hline}

{p2colset 5 16 18 2}{...}
{p2col :{hi:r2o} {hline 2}} r2o ordinal explained variation statistic {p_end}
{p2colreset}{...}

{title:Syntax}

{p 8 15 2}
{cmd:r2o}  [,{it:}{cmd:TRustme}] [{it:}{cmd:NOmarg}]   {p_end}

{synoptline}
{title:Description}

{pstd} {cmd:r2o} calculates the ordinal explained variation statistic (i.e., Rē) described by Lacy (2006), which
is used to summarize the fit of a regression model for an ordinal response. It rests on an ordinal variation
measure that entails no assumptions about intercategory distances or distributional form.  This measure is valid regardless 
of the method used to estimate the model, and was shown to outperform various pseudo-Rē measures in estimating the value of 
the true Rē for a regression model for an underlying continuous response, even though its sense does not require such. 
{cmd:r2o} is to be used after a relevant categorical response model has been run, while the e() list is still intact. By default, the program recognizes
the following as relevant response models: {helpb ologit}, {helpb oprobit}, {helpb mlogit}, and {helpb gologit2}. However, {cmd:r2o}  
should work after any estimation command for which {helpb predict} {it: p1,..., pk} will calculate predicted probabilities, and
which follows official Stata's conventions for naming items in the e() list. {p_end}


{title:Options}
{synoptset 16 tabbed}{...}
{synopthdr}
{synoptline}
{synopt:{opt trustme}} will make the program run regardless of whether an estimation command it recognizes has just been run. {p_end}
{synopt:{opt nomarg}} suppresses the display of the marginal distribution of the response variable. {p_end}

{title:Saved results}
{pstd}

{cmd:r2o} saves the following in {cmd:r()}:
{synoptset 15 tabbed}{...}
{synopt:{cmd:r(vtot)}} total variation in the response variable {p_end}
{synopt:{cmd:r(verr)}} error variation, i.e., the conditional variation in the response {p_end}
{synopt:{cmd:r(vmodel)}} variation explained by the model {p_end}
{synopt:{cmd:r(r2o)}} ordinal explained variation statistic. {p_end}
{synopt:{cmd:r(ur2o)}} bias adjusted version of r2o. {p_end}

{title:Examples}

{psee}{cmd:}{p_end}
{psee}{cmd:. sysuse auto}{p_end}
{psee}{cmd:. ologit rep78 price foreign }{p_end}
{psee}{cmd:. ologit rep78 price foreign, nomarg }{p_end}

{title:References}

{phang}Lacy, Michael G. 2006. "An Explained Variation Measure for Ordinal Response Models With Comparisons to Other Ordinal Rē Measures." 
Sociological Methods and Research. 34:469-520. 


{title:Author}

{phang}Michael G. Lacy{p_end}
{phang}Colorado State University{p_end}
{phang}Fort Collins, Colorado{p_end}
{phang}Michael.Lacy@colostate.edu{p_end}
{phang} http://central.colostate.edu/author/mglacy/ {p_end}

{title:Acknowledgements}

{phang} My thanks go to Richard Williams for helping me with various programming issues. {p_end}