VECM Entering Intercepts and Trends

Questions and discussions on Vector Autoregressions

VECM Entering Intercepts and Trends

Postby dunt » Mon Aug 01, 2011 2:56 pm

I am running a money demand model VECM model and wish to allow for an intercept in the cointegrating vector and for a linear trend in the VAR. The closest I can get to this is the following commands which appear to allow for an intercept and trend in the VAR but do not allow for an intercept in the cointegrating vector:

source(noecho) c:\winrats\johmle.src
@johmle(lags=2,det=trend, CV=CVECTOR)
# lrealM1 lrealGDP OC
equation(coeffs=cvector) ecteq *
# lrealM1 lREALGDP OC
system(model=ectmodel)
variables LREALM1 lREALGDP OC
lags 2
ect ecteq
end(system)
estimate

Can this be amended to include an intercept in the vector?
dunt
 
Posts: 1
Joined: Sat Jul 30, 2011 12:10 pm

Re: VECM Entering Intercepts and Trends

Postby TomDoan » Tue Aug 02, 2011 12:44 pm

The more recent versions of JOHMLE (the current one is at http://www.estima.com/forum/viewtopic.php?f=7&t=239) has the option DET=RC, which restricts the constant to the cointegrating vector. By constrast, DET=CONSTANT includes an unrestricted constant, which allows/produces a trend in the variables.
TomDoan
 
Posts: 2720
Joined: Wed Nov 01, 2006 5:36 pm


Return to VARs (Vector Autoregression Models)

Who is online

Users browsing this forum: No registered users and 1 guest

cron