Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

Re: st: Solution for Autocorrelation in Lag orders for VAR

From   Muhammad Anees <>
Subject   Re: st: Solution for Autocorrelation in Lag orders for VAR
Date   Mon, 6 Feb 2012 09:17:28 +0500

Are you comparing different countries over time? Is this not the case
of panel data?

Rest the Baums advice is enough to start ahead. Most of the Time
Series studies in the Energy Economics uses sample of 35-40
observations, so checking for these studies might be helpful in this
regard. A few studies are listed below;

1. Halicioglu, F., 2009. An econometric study of CO2 emissions, energy
consumption, income and foreign trade in Turkey, Energy Policy 37,
2. Ghosh, S., 2010. Examining carbon emissions economic growth nexus
for India: a multivariate cointegration approach, Energy Policy,
38(6), 3008-3014.
3. Narayan, P.K., Singh, B., 2007. The electricity consumption and GDP
nexus for the Fiji Islands, Energy Economics 29, 1141–1150.
4. Oh, W., Lee, K., 2004. Causal relationship between energy
consumption and GDP revisited: the case of Korea 1970–1999, Energy
Economics 26, 51–59.

Sorry for the studies not specific to your study but for VAR and VECM,
a few of these studies are good enough.



On Mon, Feb 6, 2012 at 12:07 AM, Muhammad Akram <> wrote:
> Dear,
> Thanks again. I have tested the variables for stationary and cointegration. They are stationary at first differance but have unit root at levels. So in my VAR model I am using them in difference. I tried VECM but it fails to include one trend variable (technological growth) which covers affectiveness of labour. If I use 3lags and Var at difference all is fine for other two countries but for one country only LM test fails at 1st lag it works on other two lags. What would you suggest in this case? Any reference which I can use to deal with bit of autocorelation.
> Thanks again
> aasim
> ---------------------------------------- > From: > To: > Date: Sun, 5 Feb 2012 12:05:21 -0500 > Subject: re:RE: st: Solution for Autocorrelation in Lag orders for VAR > > <> > Also it gives problem for eigen values table and graph with few values outside eigen circle.So I have to stick to this lag length. Do you have any idea to solve this issue in STATA. > > In Stata (not STATA!) this evidence from the eigensystem suggests that some of your variables fail to satisfy the requirements for dynamic stability of the VAR, possibly because they are unit root (I(1)) processes. It is difficult to test them for stationarity given only 30 obs. but if other research using these variables has focused on their differences rather than their levels, I would heed it. > > Kit > > > Kit Baum | Boston College Economics & DIW Berlin | > An Introduction to Stata Programming |!
>  ml > An Introduction to Modern Econometrics Using Stata | > > > * > * For searches and help try: > * > * > *
> *
> *   For searches and help try:
> *
> *
> *


Muhammad Anees
Assistant Professor/Programme Coordinator
COMSATS Institute of Information Technology
Attock 43600, Pakistan

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index