I have to admit that the GRETL package was proven better beyond any expectations.
During my late work, an extensive study of water consumption in Athens, I became familiar
with a variety of statistical packages including SPSS (statistical with R-plug-in
functions), GRETL (V 1.10.1), Past, Rapid miner and Project-R.
In the field of Non-Linear Statistics, my conclusions could me summarized as following:
SPSS and Gretl seem to be of the same functionality in NLS. The results ware comparable
and were equally well presented (graphs and summary statistics).
In term of convergence, SPSS seems to be more "manageable" because it was easily
possible to define "Range" in constant and coefficients. Gretl was more
sensitive in initial values, although later, I realized that is was possible to force
constrains. On the other hand, Gretl was much faster, allowing user to run many
experiments in a sort time. I will not be more detailed about the models of NLS methods
because I was mainly interested in results and stability of model.
In terms of graph creation, although SPSS has more features, it was really slow, resulting
) have also Non-linear regression, very easy to use,
but with less documentation than the previous packages.
In general, I found very useful the various tests of models' stability, like CUSUM and
the very compact menu of "MODEL" in Gretl. Just with few clicks I was able to
test many models that in other packages it was a time consuming procedure.
I mentioned Past package only for the fact that includes same "exotic" feature
like wavelet scalograms.
Unfortunately I don't have in hand an English version of my work yet, but I will send
you a link as soon as it will available.
H D Mechanical Engineer, NTUA
MS IEOR Columbia Univerity
MS Computer Science, Hellenic Open University
Athens Water Supply and Sewerage Company (EYDAP S.A.)
From: gretl-users-bounces(a)lists.wfu.edu [mailto:email@example.com] On
Behalf Of Agustin Alonso
Sent: Monday, October 17, 2016 9:46 PM
Subject: [Gretl-users] Nonlinear Time Series
During the past 2 years I have not used Gretl, because of that, I would like to know if
the package has now the facilities for studing nonlinear time series.
Thanking you for your attention, I am
Prof. of Econometrics
Real Centro Universitario "Escorial-Maria Cristina"
San Lorenzo de El Escorial(Madrid) Spain
νέα εφαρμογή EydApp | Η επόμενη μέρα στην Εξυπηρέτηση
με φροντίδα για το νερό
ΓΝΩΣΤΟΠΟΙΗΣΗ-ΔΗΛΩΣΗ Αυτό το μήνυμα, ή/και κάθε συνημμένο αρχείο του, είναι εμπιστευτικό
και απόρρητο. Σε περίπτωση, που περιέλθει σε εσάς από λάθος, χωρίς να είστε ο παραλήπτης
(ή πρόσωπο υπεύθυνο να το παραδώσει σε αυτόν), παρακαλούμε να το διαγράψετε αμέσως από το
σύστημά σας, χωρίς να το αντιγράψετε/προωθήσετε σε τρίτους/ανοίξετε, καθώς και να
ειδοποιήσετε άμεσα τον αποστολέα του. Επίσης, η ΕΥΔΑΠ Α.Ε. δεν αναλαμβάνει καμία ευθύνη,
για οιαδήποτε ζημία, προκαλέσει αυτό το μήνυμα, στο σύστημα ή τα δεδομένα σας.
CONFIDENTIALITY WARNING-DISCLAIMER This message and/or its attachments is confidential and
privileged information. If you are not the intended recipient(s) you should not copy it,
re-transmit it, use it, or disclose its contents, but should return it immediately to the
sender and delete it from your system. It is also noted that EYDAP S.A. shall not be
responsible, for any loss, disruption or damage to your data or computer system that may
occur while using data contained in, or transmitted with, this message.
PThink before you print - save water!