Harris and Sollis cite Johansen's suggestion in choosing which of the three
models (Models 2, 3, and 4) in constructing a VEC model, one needs "to test
the joint hypothesis of both the rank order and the deterministic
components, based on the so-called Pantula principle." (p. 134-5)
"All three models are estimated and the results are presented from the most
restrictive alternative (i.e., r=0 and Model 2) through to the least
restrictive alternative (i.e., r=n-1 and Model 4). The test procedure is
then to move through from the least restrictive model and at each stage to
compare the trace test statistic to its critical value and only stop the
first time the null hypothesis is not rejected."
Then, using, UK real demand for money data, they present their Table 5.5
that lists the trace statistics for three different models at each rank
level and the statistic for which the first time the null is not rejected
is marked with a star.
I believe adding this procedure under Time Series section of the Model menu
for those of us who are not good at coding will be very helpful. However,
for its urgency for me, if someone could help me in coding this procedure
using one of gretl's sample data, I'll really appreciate it.
P.S. I tried but was not able to use "Model table" in icon view for this
Harris, R and R. Sollis (2003) Applied Time Series Modelling and
Johansen, S. (1992) Determination of cointegration rank in the presence of
a linear trend, Oxford Bulletin of Economics and Statistics, 54, 383-397.
Show replies by date