Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | David Torres <writeon4truth2@msn.com> |
| To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
| Subject | RE: st: Graphing a function (setting coefficients at their means) |
| Date | Mon, 30 Dec 2013 15:39:35 -0500 |
Okay, Nick, So if I can't really graph my equation using -twoway function-, is there a way to do so using margins? I assumed that it's impossible since I was using a nonconventional estimator (-xtivreg-). For a bit more context, I have a time-varying binary endogenous predictor that I'm instrumenting for. This endogenous variable is interacted with dummies for race, but all appear in the parenthetical portion of -xtivreg-. I want to show the marginsplot for race, taking into account the interactions, but I'm afraid the -xtivreg- is not recognizing my endogenous variable as time-varying. Maybe I'm doing something wrong with the IV estimation. ?? Diego -- David Diego Torres, PhD NSF SBE Postdoctoral Research Fellow Houston Education Research Consortium Rice University 6100 Main Street, MS-28 Houston, TX 77005 Phone: 713-348-2984 Email: ddtorres at rice dot edu ---------------------------------------- > Date: Mon, 30 Dec 2013 14:05:26 +0000 > Subject: Re: st: Graphing a function (setting coefficients at their means) > From: njcoxstata@gmail.com > To: statalist@hsphsun2.harvard.edu > > If I understand you correctly, you need to do the manipulations before > and outside -twoway function-. -twoway function- contains no > intelligence to do with previously estimated models. You can, > naturally, refer to r-class results and e-class results in calls to > -twoway function-, but I am confident in speculating that -twoway > function- has no hidden options to make marginal calculations easier. > > Nick > njcoxstata@gmail.com > > > On 30 December 2013 12:14, David Torres <writeon4truth2@msn.com> wrote: >> Statalisters, >> >> I'm using the -twoway function- command to display the results of a regression equation. Is there a way to set at their means those coefficients that are not of interest in the command line? >> >> Just in case you're wondering why I'm not using -margins- to get predicted probs, I can't because I'm using the IV estimator, which presents problems for a variable of interest to be graphed. >> >> Cheers, >> Diego >> Sent from my iPhone >> * >> * For searches and help try: >> * http://www.stata.com/help.cgi?search >> * http://www.stata.com/support/faqs/resources/statalist-faq/ >> * http://www.ats.ucla.edu/stat/stata/ > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/