On Thu, 5 Jul 2012, JOSE FRANCISCO PERLES RIBES wrote:

> I'm doing a unit root test ADF with Gretl on a series of tourism market
> share of Spain specified with constant and trend.
> By comparing the results with Eviews or R (package fUnitRoot) I get the same
> t-statistic, but although both programs indicate that the critical values
> ​​are McKinnon (1996) MacKinnon, J. G. (1996) "Numerical distribution
> functions for unit root and cointegration tests", Journal of Applied
> Econometrics 11: 601-618.
> p-values ​​of the test are very different in either case .
> Gretl: t = -3.62 p-value 0.02 asymptotic
> Eviews t = -3.62 p-value (one-sided) = 0.04 which is the same value
> obtained in R.

You should find that if you do a non-augmented Dickey-Fuller test 
(no lagged differences) the P-values given by gretl agree with those 
from R's fUnitRoots package. If you run an augmented test, gretl 
automatically gives the asymptotic P-value, while it appears that 
fUnitRoots is giving the finite-sample value for the sample size 
used (and I suppose Eviews is doing the same). I verified this on 
some examples, using MacKinnon's urcdist program.

I believe gretl is doing the right thing here. In his 1996 JAE 
article MacKinnon says, "Since the finite-sample P-values are valid 
only for non-augmented Dickey-Fuller tests, it is probably wise to 
ignore them for ADF tests..."

Allin Cottrell

Reply via email to