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

st: Re: About Stata


From   Kit Baum <kitbaum@mac.com>
To   Miguel Angel Alcantara <malcantara@lamolina.edu.pe>
Subject   st: Re: About Stata
Date   Fri, 12 Aug 2005 17:25:38 -0400

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/



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