{smcl}
{* 04Aug2009}{...}
{cmd:help paragr}

{hline}

{title:Title}

{p2colset 5 15 22 2}{...}
{p2col :{hi:  paragr} {hline 2}}Parallel graphing of a coefficient across different equations{p_end}

{marker s_Syntax}
{title:Syntax}

{p 4 8 6}
{cmdab:paragr} {it: variable} [, {it:options}]

{marker s_Description}
{title:Description}

{p 4 8 6}
{cmd:paragr} provides a fast and easy way to compare a coefficient across different equations 
within an estimation.

{p 4 8 6}
It can used to visualize the parallel assumption of ordered logit or 
the equality of coefficients for quantile regression, etc.

{p 4 8 6}
By default {cmd:paragr} plots a scatter graph. A quadratic fitted line (qfit) can be superimposed.

{p 4 8 6}
Twoway graph options can be specified for the entire graph, or separately for the scatter graph 
or for the qfit graph.

{marker s_Options}
{title:Options}

{dlgtab:Main}

{p 4 12 6}{opt scat:ter( )} standard options for {help scatter graph}. Use this to set color, range, etc {p_end}

{p 4 12 6}{opt qfit} a quadratic fitted line is superimposed. {p_end}

{p 4 12 6}{opt qfit( )} standard options for {help twoway_qfit:qfit graph}. Use this to set color, range, etc {p_end}

{p 4 12 6}{opt gen( )} a prefix for saving the plot points as variables.{p_end}

{p 4 12 6}{help twoway_options:click here} for a list of {help twoway} options, including:{p_end}

{p 8 12 6}{opt nodraw} suppress the display of graph{p_end}

{p 8 12 6}{opt saving( )} save as a Stata graph{p_end}

{p 8 12 6}{opt name(name[, replace])} specifies the name of the graph

{marker s_0}
{title:Examples}

{p 4 8 6}* reg3{p_end}
{p 4 8 6}{stata sysuse auto, clear}{p_end}
{p 4 8 6}{stata reg3 (price weight rep78) (price weight rep78 mpg) (price weight rep78 mpg headroom) (price weight rep78 mpg headroom length)}{p_end}
{p 4 8 6}{stata paragr rep78, xtitle(Equation) gen(reg3_) saving(mygraph)}{p_end}


{p 4 8 6}* sqreg {p_end}
{p 4 8 6}{stata sysuse auto, clear}{p_end}
{p 4 8 6}{stata sqreg price mpg rep78 headroom length, q(.10 .20 30 .40 .50 .60 .70 .80 .90) reps(25)}{p_end}
{p 4 8 6}{stata paragr rep78, qfit}{p_end}


{p 4 8 6}* Unconstrained gologit {p_end}
{p 4 8 6}{stata sysuse auto, clear}{p_end}
{p 4 8 6}{stata egen money=cut(price), group(5)}{p_end}
{p 4 8 6}{stata gologit money headroom mpg weight length turn displacemen}{p_end}
{p 4 8 6}{stata paragr displacement}{p_end}


{p 4 8 6}* constrained gologit2 from ssc (need Stata 8.2 or better) {p_end}
{p 4 8 6}{stata sysuse auto, clear}{p_end}
{p 4 8 6}{stata egen money=cut(price), group(5)}{p_end}
{p 4 8 6}{stata gologit2 money headroom mpg weight length turn displacement, auto}{p_end}
{p 4 8 6}{stata paragr weight, qfit ytitle(odds ratios) xtitle(cut)}{p_end}


{title:Remarks}

{p 4 12 6}{opt qfit} requires at least two equations.{p_end}

{p 4 12 6}Coefficient values goes on the y-axis.{p_end}


{title:Acknowledgement}

{p 4 12 6}Nick Cox provided detailed and helpful suggestions.{p_end}


{title:Author}

{p 4 8 6}Roy Wada{p_end}
{p 4 8 6}roywada@hotmail.com{p_end}