Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, is already up and running.

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: confidence intervals for IRFs

From   ulio Glogowsky <>
Subject   Re: st: confidence intervals for IRFs
Date   Wed, 19 Sep 2012 15:19:26 +0200

Sorry! I'm going to reformulate the problem:

I estimated a system of equations with a vector autoregressive model
for two variables y1 and y2 and one lag.
What I ultimately want to do is to construct confidence intervals for
impulse response functions of this model. One way to do this is by
Monte Carlo Integration. Doan explains in the "RATS User's Guide" how
this can be done.

When I try to write a program with the same procedure in STATA I start
by calculating the covariance matrix of the estimated system of
equations (here: "sigma" with dimension 2x2). This is not a problem.
Now the inverse of sigma is decomposed by cholesky (lets call the new
sigma sigmad) and based on sigmad and the degrees of freedom
covariance matrices are drawn from an inverse Wishart Distribution.
So, in order to use the same procedure in Stata, I would also have to
draw from the inverse Wishart distribution. However I cannot find a
way to do this.

Is there a way to draw from the Wishart distribution in Stata similar
to drawnorm()?

2012/9/19 Maarten Buis <>:
> On Wed, Sep 19, 2012 at 10:29 AM, ulio Glogowsky wrote:
>> I want to use the method outlined by Doan (1990) to construct
> The Statalist FAQ exists to help you ask answerable questions. Please
> use that service.
> -- Maarten
> ---------------------------------
> Maarten L. Buis
> Reichpietschufer 50
> 10785 Berlin
> Germany
> ---------------------------------
> *
> *   For searches and help try:
> *
> *
> *
*   For searches and help try:

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