Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | tom smith <tom_smith_tomm@hotmail.com> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | st: Using GARCH to forecast |
Date | Sat, 24 Nov 2012 16:26:16 +0000 |
Dear all, I have a question regarding the GARCH estimation. I would like to evaluate the predictions of a GARCH estimation on the basis of a part of my own sample. The way I did this was: gen d1=1 if q>=yq(1990,2) & qtr<=yq(2012,2) replace d1=0 if d1==. To generate a dummy that defines the part I want to use to create the model and the part to check it. arch X if d1==0, arch(1) garch(1) Then I apply the GARCH on my variable X But my question is now how can I validate this model? because when you use an ARIMA model you could just use: predict forecast if val2==1 And then compare the forecast with the variable. Sorry but I'm new to TS with stata... Thanks, Tom * * 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/