There seems to be a bug in gretl_export and / or mread functions, at least when one uses Octave.
Check the following script:
combj = nchoosek(1:6,3)
combj1 = mread("@dotdir/combj1.mat")
One can retrieve the original matrix by this operation, but I guess that is not intended:
combj2 = mshape(vec(combj1'),rows(combj1),cols(combj1))
Keep the excellent work.
Hi,I've been using gretl for different kinds of models and I have a question about the ARIMA models. I use the graphic interface and not the console, because I haven't had time to learn the commands. After calculating the models with several lags on the dependent variable I attempted to use the parameters to replicate the fitted values using Excel. My question is how the fitted values are calculated, because I don't know how the first values are calculated, since there is no data for the lagged variable.
I figured that if, for example, the model I calculated uses the third and fourth lag of the dependent variable, then the first and second fitted values set the third and fourth lag to be equal to zero, so that the parameter for the third and fourth lag is not used. But what happens at the third fitted value?
Thank you for your help and for this great program. Please let me know if you need any more information.
I'm still trying to get a feeling for the dpanel gmm estimators.
When estimating this xtabond2 statement from Roodman (2006/2008) for
"xtabond2 n L.n L(0/1).(w k) yr*, gmmstyle(L.(n w k)) ivstyle(yr*,
equation(level)) robust small"
dpanel 1; n const w w(-1) k k(-1) ; \
GMM(n,2,8) GMM(w,2,8) GMM(k,2,8) \
GMMlevel(w,1,1) GMMlevel(k,1,1) --time --sys
I came across a question concerning the Sargan/Hansen test of overid.
In the gretl-guide on p.152 it is stated that "Specifically, xtabond2
computes both a "Sargan
test" and a "Hansen test" for overidentification, but what it calls the
Hansen test is what DPD and
gretl call the Sargan test."
The Hansen test in this example does not reject the validity of the
instruments while the Sargan does.
"Sargan test of overid. restrictions: chi2(100) = 186.90 Prob > chi2 = 0.000
(Not robust, but not weakened by many instruments.)
Hansen test of overid. restrictions: chi2(100) = 110.70 Prob > chi2 = 0.218
(Robust, but can be weakened by many instruments.)"
In gretl output however the result of the Sargan test is and not the
Hansen test :
"Sargan over-identification test: Chi-square(100) = 154.808 [0.0004]"
That's was quite a surprise for me.
Perhaps it's because the test statistic is not the same and it's really
the Hansen test (cos I believe in what u documented :-)), but why then
such drastic differences?
I think that accessors for the various test-statistics and p-values
would be useful like $ar1, $ar2.
Have you already thought about incorporating a 'collapse' switch?
It would be a good feature for robustness checks.
In my humble opinion the implementation in the (difficult) existing
instrument-matrix-generation-process seems not so hard, since the rows
of the standard matrix just move to the right if there is a zero entry
to the right.
The GMM() function does not support the use of lists right now. Could
this be improved?