{smcl} {* *! version 1.0.2 Nicolai T. Borgen 15june2022}{...} {cmd:help rqrplot} {hline} {title:Title} {p2colset 5 16 19 20}{...} {p2col :{hi:rqrplot} {hline 2}}Graphing quantile regression coefficients after RQR{p_end} {p2colreset}{...} {title:Syntax} {p 8 13 2} {cmd:rqrplot} [{cmd:, } {opth bopts(string)} {opth ciopts(string)} {opth twopts(string)} {opt level(#)} {opth bootstrap(string)} {opt nodraw} {opt notabout} {opt noci}] {marker options}{...} {synoptset 27 tabbed}{...} {synopthdr :options} {synoptline} {p2coldent : {opth bopts(string)}}allows for the customizing the display of the coefficients. The default is solid line graph. See {it:twoway {help twoway##options:options}} for other line options.{p_end} {p2coldent : {opth ciopts(string)}}allows for customizing the confidence intervals. The default is area plot with opacity set at 40%. See {it:twoway {help twoway_options##options:options}} for other options.{p_end} {p2coldent : {opth twopts(string)}}allows for customizing the overall graph, including title and labels. See {it:twoway_options {help twoway##options:options}} for various options.{p_end} {p2coldent : {opt level(#)}}decides the confidence level for the confidence intervals, where # is any number between 10.00 and 99.99. The default is 95% confidence interval.{p_end} {p2coldent : {opth bootstrap(string)}}requests normal-approximation bootstrap CIs ({cmd:bootstrap(normal)}), percentile bootstrap CI ({cmd:bootstrap(percentile)}), or bias-corrected bootstrap CI ({cmd:bootstrap(bc)}). The default is normal-approximation when {helpb rqr} is estimated with the {helpb bootstrap} prefix.{p_end} {p2coldent : {opt nodraw}}suppresses the display of the {cmd:twoway} plot.{p_end} {p2coldent : {opt notabout}}suppresses the display of the result matrix.{p_end} {p2coldent : {opt noci}}plots the coefficients without confidence intervals.{p_end} {synoptline} {p2colreset}{...} {marker weights}{...} {pstd} {title:Description} {pstd} {cmd:rqrplot} is a {helpb rqr} postestimation command that effortless plots quantile regression coefficients and their confidence intervals. It visualizes the coefficients and the confidence intervals based on the current estimation results from the {helpb rqr} model. {pstd} The {helpb rqrplot} postestimation command only works after the {helpb rqr} command. {pstd} See {browse "https://osf.io/preprints/socarxiv/4vquh": Borgen, Haput, and Wiborg (2021b)} for descriptions and examples of the {cmd:rqr} and {helpb rqrplot} commands. {title:Examples: Basic usage} {pstd} Setup{p_end} {phang2}{cmd:. webuse nlswork} {pstd} Estimate treatment effects for quantiles 0.03 to 0.097 at steps of .02 using {helpb rqr}.{p_end} {phang2}{cmd:. rqr ln_wage union, quantile(.03(.02).97) controls(year c.grade##c.grade south i.ind_code)} {pstd} Plot treatment coefficients using defaults. {p_end} {phang2}{cmd:. rqrplot} {title:Examples: Customize graph layout} {pstd} Change line pattern from solid to dash. {p_end} {phang2}{cmd:. rqrplot, bopts(lpattern(dash))} {pstd} Change colors of line and CIs. {p_end} {phang2}{cmd:. rqrplot, bopts(color(red)) ciopts(color(red%40))} {pstd} Change line plot to connected plot. {p_end} {phang2}{cmd:. rqrplot, bopts(recast(connected))} {pstd} Change markers of connected plot. {p_end} {phang2}{cmd:. rqrplot, bopts(recast(connected) msymbol(square) msize(vsmall))} {pstd} Change area CIs to line CIs. {p_end} {phang2}{cmd:. rqrplot, ciopts(recast(rline))} {pstd} Change confidence level, here with 70% CIs. {p_end} {phang2}{cmd:. rqrplot, level(70)} {pstd} Add overall title and change the default y-titles. {p_end} {phang2}{cmd:. rqrplot, twopts(title(Union wage effects) ytitle(QTE))} {pstd} Remove grid lines. {p_end} {phang2}{cmd:. rqrplot, twopts(ylabel(,nogrid) xlabel(,nogrid))} {pstd} Suppress the confidence intervals. {p_end} {phang2}{cmd:. rqrplot, noci} {pstd} Names for graph are specified in {opt twopts(string)}. {p_end} {phang2}{cmd:. rqrplot, twopts(name(union_effects, replace))} {pstd} Combine some of the different options mentioned above. {p_end} {phang2}{cmd:. rqrplot, bopts(recast(connected) color(red) msymbol(oh)) ciopts(color(red%40)) twopts(title(Union wage effects) ytitle(QTE))} {title:Examples: Bootstrapped confidence intervals} {pstd} To get bootstrapped CIs, we first need to run {cmd:rqr} with the {cmd:bootstrap} prefix. To save time, we will restrict the anlysis to 33-36-year-olds and only run 20 reps. {p_end} {phang2}{cmd:. bootstrap, reps(20): rqr ln_wage union if inrange(age,33,35),, quantile(.03(.02).97) controls(year c.grade##c.grade south i.ind_code)} {pstd} When {cmd:rqr} is estimated using the {cmd:bootstrap} prefix, the default in {cmd:rqrplot} is to plot the normal-approximation CIs. {p_end} {phang2}{cmd:. rqrplot} {pstd} To get percentile-based CIs:{p_end} {phang2}{cmd:. rqrplot, bootstrap(percentile)} {title:Stored results} {cmd:rqr} stores the following in {cmd:r()}: {synoptset 20 tabbed}{...} {synopt:{cmd:r(plotmat)}}matrix containing quantile, coefficient, standard errors, and upper and lower confidence intervals{p_end} {p2colreset}{...} {title:Version requirements} The {cmd:rqrplot} command requires Stata 12.0 or later. {title:Reference} {p 4 8 2} {browse "https://osf.io/preprints/socarxiv/42gcb/": Borgen, Haupt, and Wiborg (2021a)} A New Framework for Estimation of Unconditional Quantile Treatment Effects: The Residualized Quantile Regression (RQR) Model. {it:SocArXiv}. doi:10.31235/osf.io/42gcb{p_end} {p 4 8 2} {browse "https://osf.io/preprints/socarxiv/4vquh": Borgen, Haupt, and Wiborg (2021b)} Flexible and fast estimation of quantile treatment effects: The rqr and rqrplot commands {it:SocArXiv}. doi:10.31235/osf.io/4vquh{p_end} {title:Authors} {p 4 4 2} Nicolai T. Borgen, University of Oslo{break} n.t.borgen@isp.uio.no{p_end} {p 4 4 2} Andreas Haupt, Karlsruhe Institute of Technology{break} andreas.haupt@kit.edu{p_end} {p 4 4 2} Øyvind Wiborg, University of Oslo{break} o.n.wiborg@sosge.uio.no{p_end} {p 4 4 2} Thanks for citing this software in one of the following ways: {p_end} {p 8 8 2} {browse "https://osf.io/preprints/socarxiv/42gcb/": Borgen, NT., A. Haupt, and ØN. Wiborg (2021).} A New Framework for Estimation of Unconditional Quantile Treatment Effects: The Residualized Quantile Regression (RQR) Model. {it:SocArXiv}. doi:10.31235/osf.io/42gcb{p_end} {p 8 8 2} {browse "https://osf.io/preprints/socarxiv/4vquh": Borgen, Haput, and Wiborg (2021b)} Flexible and fast estimation of quantile treatment effects: The rqr and rqrplot commands {it:SocArXiv}. doi:10.31235/osf.io/4vquh{p_end}