Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Lin, Bill" <bill.lin.79@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: forecast after reg |
Date | Tue, 26 Oct 2010 22:03:58 -0400 |
Thanks alot, Robert! On Tue, Oct 26, 2010 at 9:09 PM, Robert A Yaffee <bob.yaffee@nyu.edu> wrote: > Bill, > You probably have a conditional forecast to perform. This means that you have to forecast all > of your weakly exogenous variables prior to forecasting your endogenous variable. Without > doing that first, you will be unable to generate the endogenous forecast. > Cheers, > Robert > > Robert A. Yaffee, Ph.D. > Research Professor > Silver School of Social Work > New York University > > Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf > > CV: http://homepages.nyu.edu/~ray1/vita.pdf > > ----- Original Message ----- > From: "Lin, Bill" <bill.lin.79@gmail.com> > Date: Tuesday, October 26, 2010 3:52 pm > Subject: Re: st: forecast after reg > To: statalist@hsphsun2.harvard.edu > > >> Thanks very much. Sorry I didn't make it clear. But I did tsset the >> data before lag and use tsappend, add(10) before forcast. It didn't >> work, so I am still searching for a workable way. >> >> >> On Tue, Oct 26, 2010 at 1:55 PM, Robert A Yaffee <bob.yaffee@nyu.edu> >> wrote: >> > Bill, >> > You have to use the tsappend, add(10) >> > command before you forecast but after you have >> > tsset your date-time variable. >> > - Bob >> > >> > >> > Robert A. Yaffee, Ph.D. >> > Research Professor >> > Silver School of Social Work >> > New York University >> > >> > Biosketch: http://homepages.nyu.edu/~ray1/Biosketch2009.pdf >> > >> > CV: http://homepages.nyu.edu/~ray1/vita.pdf >> > >> > ----- Original Message ----- >> > From: "Lin, Bill" <bill.lin.79@gmail.com> >> > Date: Tuesday, October 26, 2010 1:22 pm >> > Subject: st: forecast after reg >> > To: statalist@hsphsun2.harvard.edu >> > >> > >> >> Dear All >> >> >> >> I have a linear model with lags for a time series data, ex. >> >> y=a+b*L1(y)+cX. After the estimation, I want to do a 10 period ahead >> >> forecast. I tried to add time points in the data set and to use >> >> predict for this purpose, but it didn't work. Can any one give some >> >> suggestions? Thanks a lot, >> >> >> >> >> >> Bill >> >> * >> >> * For searches and help try: >> >> * http://www.stata.com/help.cgi?search >> >> * http://www.stata.com/support/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/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/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/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/statalist/faq * http://www.ats.ucla.edu/stat/stata/