Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: granger causality and OLS


From   <raffaella_00@yahoo.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: granger causality and OLS
Date   Fri, 21 Nov 2003 15:47:45 -0800 (PST)

Never mind - I got it.

> X               Y
> 77.76959	87
> 80	        87
> 60	        87
> 60	
> 60	        86
> 60	        86.99999
> 70	
> 70	        92
> 70	

> 1.  Why does gcause use fewer observations (3
> instead
> of the 6 available)?

Because it is a system of equations, and therefore it
only keeps the observations where also the lagged
variable of Y is nonmissing.

> If I run gcause, I get the following output:
> 
> . gcause X Y, lags(1) reg
> Granger causality test                        
> Sample:
> 2002m11 to 2003m3
>                                                     
>  
>           obs = 3
> H0: Y does not Granger-cause X
> 
>                F( 1, 1) =    0.00
>                Prob > F =   1.0000
> 
>                 chi2(1) =    0.00      (asymptotic)
>             Prob > chi2 =  1.0000      (asymptotic)

> 2.  Why does it allow to test granger causality if Y
> is dropped in the regression?

The test rejects Granger causality, which is what we
expect since in the three observations used

  > X               Y
  > 77.76959	87
--> 80	        87
--> 60	        87
  > 60	
  > 60	        86
--> 60	        86.99999
  > 70	
  > 70	        92
  > 70	

Y is always 87, and therefore it does not explain
anything because it is collinear with the constant.

Raffaella


__________________________________
Do you Yahoo!?
Free Pop-Up Blocker - Get it now
http://companion.yahoo.com/
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index