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]

st: questions on panel VAR

From   Costas Lambrinoudakis <>
To   Statalist <>
Subject   st: questions on panel VAR
Date   Sat, 23 Mar 2013 12:46:20 +0200

Dear all,

I am using the panel VAR code pvar.ado created by Inessa Love from
World Bank, along with sgmm.ado and Helm.ado. I would be grateful if
anyone could help me with the following questions:

1. It is advised to time-demean variables before doing the Helmert
procedure. In the pvar command should I use the original
(untransformed) variables or time-demeaned variables?

2. What should I do, if I want to use pvar without time fixed effects?

3. How do I include exogenous variables in the model? The syntax is
pvar varlist [if exp], [lag(p) options], but all variables in the
varlist are treated as endogenous. How am I supposed to include
exogenous variables?

4. Is there a way to test if the errors in the panel VAR are serially
correlated? (something like the estat abond with the artests(3) option
in the xtabond and xtdpsys commands)

5. Is there a way to test for the optimal number of lags in the panel
VAR and for Granger causality?

6. I noticed that in Inessa Love's 2006 paper (Tables 4 and 5) the
authors are not using the usual t-stat critical values to derive
statistical significance at the 1% and 5% level. Has anyone any idea
which critical values they are using?

Finally does anyone know any other way to run panel VAR?

Any help would be greatly appreciated!

*   For searches and help try:

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