Teresa,
Have you tried double differencing and demeaning those variables? If they are
not essential to the model and their inclusion does not improve the information criteria,
they need not be included.
Regards,
Bob Yaffee
Robert A. Yaffee, Ph.D.
Research Professor
Silver School of Social Work
New York University
Senior Research Scientist,
NSF/DRU # 0826983
----- Original Message -----
From: Teresa Linz <[email protected]>
Date: Monday, July 28, 2008 5:56 am
Subject: st: non stationary time series with SUR
To: [email protected]
> Dear stata subscribers,
>
> I have a problem with some time series analysis. I am
> estimating 45 monthly time series with seemingly unrelated
> regression. I have five independent variables which are
> estimated in a cost function as a translog functional
> form. However two of the variables are non stationary
> (also the transformed ln variables are non stationary).
> what can I do about that. Do I have to include them in a
> differenced form in the SUR model? How do I do that?
> The command I am using right now looks like that:
> sureg (Sw = lnPw lnPe lnPl lnPc lny) (Se = lnPw lnPe lnPl
> lnPc lny) (Sl = lnPw lnPe lnPl lnPc lny) (Sc = lnPw lnPe
> lnPl lnPc lny)(lnTC = lnw lne lnl lnc lnd lny lnww lnee
> lnll lncc lndd lnyy lnwe lnwl lnwc lnwd lnel lnec lned
> lnlc lnld lncd lnwy lney lnly lncy lndy), constraints(1 2
> 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
> 25 26 27 28 29 30 31) isure corr nolog.
> Here lnl and lnc are non stationary as well as all
> interrelated variables.
>
> It would be nice to get an answer about this. Thank you
> very much.
>
> Greetings,
> Teresa
>
>
>
>
> ______________________________
>
> Teresa Linz, M.Sc.
>
> Junior Researcher
>
> Center for Development Research
>
> Walter-Flex-Str. 3
>
> 53113 Bonn
>
> Germany
>
> Tel.: +49 228/73 4962
> Fax: +49 228/73 1869
> Internet: www.zef.de
> *
> * 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/
