RE: st: RE: Different results for Fixed and Random Effects models Using STATA and SAS

[Talal <talalesm@yahoo.com>]

Dear Nick,

Thanks again for your comment. I need to validate my STATA result before I can proceed with my analysis (PhD Research), SAS gives just slightly different results for Random Effect but major one for Random Effect. and it Hausman Test results is really Odd.

I will be very thankful if you can lead me to a person or contact who can help me.

Regards
Talal

--- On Tue, 11/5/10, Nick Cox <n.j.cox@durham.ac.uk> wrote:

> From: Nick Cox <n.j.cox@durham.ac.uk>
> Subject: RE: st: RE: Different results for Fixed and Random Effects models Using STATA and SAS
> To: statalist@hsphsun2.harvard.edu
> Date: Tuesday, 11 May, 2010, 10:16
> Thanks for this. Those with expert
> knowledge of knowledge of fixed and random effects models
> (not me) should be in a better position to comment or to ask
> for specific extras.  
> 
> You should probably be using an extra dummy for North-East
> England.... 
> 
> Nick 
> n.j.cox@durham.ac.uk
> 
> 
> Talal
> 
> Dear Nick,
> 
> Thanks for your comment and sorry for my shallow
> explanations of the problem. Below is the command and result
> comparison between SAS and STATA.
> I have slightly unbalaced data for 4 regional areas of GB.
> 
> 
> a-    For Fixed Effect:
> 
> SAS:
> 
> proc reg data=chapter3.all_area;
> Model Ln_qdt = Ln_qdt2 Ln_vkm Ln_income Ln_F
> deregulation_dummy Time_Trend London_dummy Mets_ dummy
> Scotland_ dummy Wales_ dummy;
> test  London_dv = Mets_dv = Scotland_dv = Wales_dv = 0
> ;
> run;
> 
> 
> STATA:
> 
> regress   Ln_qdt Ln_qdt2 Ln_vkm Ln_income
> Ln_F deregulation_dummy Time_Trend London_dummy Mets_ dummy
> Scotland_ dummy Wales_ dummy
> 
> 
> 
> 
> b-    For Random Effect:
> 
> SAS:
> 
> proc panel data=chapter3.all_area;
> ID area year;
> Model Ln_qdt = Ln_qdt2 Ln_vkm Ln_income Ln_F
> deregulation_dummy Time_Trend / RANONE BP VCOMP=WK
> ;
> run;
>                
>         
> 
> STATA:
> 
> iis area
>  xtreg qdt Ln_qdt2 Ln_vkm Ln_income Ln_F deregulation_dummy
> Time_Trend, re theta
> 
>     SAS       
> 
> Model    FE   
> RE        FE (STATA)
> RE(STATA)
>      Coeff.   
> Coeff.       
> Coeff.    Coeff.
>             
>         
> Ln F    -.108   
> -0.09892       
> -.108    -0.06321
> Ln VKM    .114   
> 0.135992       
> .115    0.082707
> Ln Income    -.560   
> -0.52503       
> -.566    -0.297
> Ln Qdt-1    .695   
> 0.747298       
> .692    0.924061
> Der. DV    -.046   
> -0.04975       
> -.047    -0.05055
> TT    .011   
> 0.010877       
> .011    0.00887
>             
>         
> Mets    .196   
>        
> 0.198    
> Scot    .153   
>        
> 0.154    
> Wales    -.023   
>        
> -0.023    
> constant   
> 5.999             
> 5.908     
>             
>         
> F           
> 3624.282       
> 3618.020    
> R2 (Adj.)    .997   
> 0.973       
> 0.9969    0.9967
> Durbin-Watson   
> 1.703                    
>             
>         
> (Incremental) F    5.57
> (0.0015)    5.57 (0.0015)    
> Breusch Pagan Test    0.00
> (0.9781)    0.00 (0.9779)    
> Hausman Test    2.34
> (0.8859)    20.16 (0.0026)   
> 
> 
> 
> --- On Mon, 10/5/10, Nick Cox <n.j.cox@durham.ac.uk>
> wrote:
> 
> > From: Nick Cox <n.j.cox@durham.ac.uk>
> > Subject: st: RE: Different results for Fixed and
> Random Effects models Using STATA and SAS
> > To: statalist@hsphsun2.harvard.edu
> > Date: Monday, 10 May, 2010, 10:34
> > This is only a small distance away
> > from "I got different results from different programs
> and
> > don't understand why". People who know about these
> commands
> > need to see exactly what you typed in both programs so
> that
> > they can be sure that the commands are exactly
> equivalent.
> > They also would find it difficult to comment unless
> your
> > results are phrased in terms of datasets that everyone
> can
> > access. 
> > 
> > Nick 
> > n.j.cox@durham.ac.uk
> > 
> > 
> > Talal
> > 
> > I have estimated Fixed and Random Effects models for
> panal
> > data which are slightly unbalanced using SAS and
> STATA
> > 
> > the two softwa estimated totally different parametres
> for
> > Ranndom Effects Model, and slightly different one for
> Fixed
> > Effects one res.
> > 
> > Is this due to diffrent estimation approches? What
> are
> > these since I have to report them on my study?
> > 
> > For Hausman test: I also got major differences between
> the
> > two softwares.
> > 
> > Is this due to diffrent estimation approches too? What
> are
> > these?
> > 
> > Is the 2 software deals with unbalanced data in
> different
> > ways?
> > 
> > I am very thankful for any answers.
> > 
> > *
> > *   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/