Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Nick Cox <njcoxstata@gmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: Trouble using -nl- |
Date | Sat, 2 Nov 2013 11:46:19 +0000 |
You don't define all the terms here, but it appears that you have a fitted model and want to use it for projection into the future. It appears that you want to add noise to deterministic predictions too. -nl- doesn't support any of that. Your problem needs customised code. Nick njcoxstata@gmail.com On 2 November 2013 09:51, Ingrid Johnsen <ingjoho@gmail.com> wrote: > Hi > > I have trouble using -nl- . > I am using Stata 12 for Windows. > > I have model consisting of three equations. > > I : y_t = a0 + a1*x_t > II : y_t = b0 + b1*y_(t-1) + sigma*e_t > III: sigma^2 = c0 + c1*cos(2*3.14* z /365) > > y is the dependent variable, x and z are the undependent variables. > I have estimated the parameter values, a0=7.68, a1=-0.00019, b0=-0.0013, > b1=0.79, c0=3.91, and c1=1.72. > I want to combine the equations into one, and make a prognosis by using the > fitted parameters. > > I think I should use -nl- to combine the equations. > But I don't understand how I can include equation III, since it is a > function modeling the residuals in equation II. > > -- > Best regards > Ingrid Johnsen > * > * 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/