st: Re: About Stata

[Kit Baum <kitbaum@mac.com>]

Dear Miguel

Presumably you have some indicator in the data set that points to the  
year of the obserrvation. Just make sure that this variable increases  
by one per year, and you can use it for tsset. That is, it could be  
1,1,2,3,4 or it could be 1997,1998,1999,2000,2001.

You cannot estimate a VAR without a number of time series  
observations. A VAR involves lags. 1998 is not the lag of 2000. If  
the consecutive obs. in your sample are 1998 and 2000, define a new  
time variable that counts by integers for those years, e.g. 1998=-3,  
2000=4 (4th period when you observe the HH). But VAR techniques with  
a total of 5 time series obs. are not really workable.

It is more efficient for you to direct such questions to Statalist.  
When I am on holiday I will not be reading Statalist, but someone who  
can help you will a.s. be available.

Kit Baum, Boston College Economics
http://ideas.repec.org/e/pba1.html


On Aug 12, 2005, at 7:13 PM, Miguel Angel Alcantara wrote:


> Professor Baum:
> Plese, allow me two consultations about Stata.
>
> 1. I have balance panel, moreless of 800 observations (households)  
> and five
> years. How I must generate time variable, before tsset?
>
> 2. It is possible to use vargranger only with two elapsed years (for
> example, 1998 and 2000)?
>
>
>
> Thanks in advance
>
> Miguel Angel Alcantara Santillan
> National Agrarian University La Molina
> Department of Economics
> Lima, Peru
> Phone: (511) 3495647, ext. 239
*
*   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/