[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

From |
Kit Baum <baum@bc.edu> |

To |
statalist@hsphsun2.harvard.edu |

Subject |
st: Re: "recursive var" |

Date |
Sat, 27 Aug 2005 09:49:52 -0400 |

What you are describing is not a VAR; it is a system of simuitaneous equations. A VAR is just what it says: an autoregression of each variable on the PAST values of itself and the other variables in the VAR. Once you have a current value of any of those variables on the RHS, you can no longer use OLS to estimate the equation. A VAR can be estimated with equation-by-equation OLS. What you describe requires instrumental variables techniques. If you want to estimate the entire system, use reg3 (three-stage least squares).

Kit Baum, Boston College Economics

http://ideas.repec.org/e/pba1.html

On Aug 27, 2005, at 2:33 AM, statalist-digest wrote:

I am trying to run a recursive VAR (vector autoregression) in which some

equation(s) of the VAR system contains the current value of other

variable(s) , not just past values of the variables.

I wonder if Stata allows for including some of the current values of

variable(s).

Anyone who is familiar with VAR in Stata, please reply.

* * 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/

- Prev by Date:
**st: Re: tsset and r451** - Next by Date:
**st: Re: [Stata 9] Convincing Stata to let me quit the program without confirmation** - Previous by thread:
**st: Re: tsset and r451** - Next by thread:
**st: simulate maximum likelihood estimation--where to construct the random variables** - Index(es):

© Copyright 1996–2016 StataCorp LP | Terms of use | Privacy | Contact us | What's new | Site index |