On Tue, 30 Mar 2010, Josephine Sudiman wrote:
Dear All,
My name is Josephine, currently doing my postgraduate research degree. I
am recently trying to learn GRETL to estimate my ACD (Autoregressive
Conditional Duration), a model which shres similar concept to GARCH.
While the latter concerns about modelling the hetero issue of mean
equation, the former is about duration. Some suggests that I have to use
Maximum Likelihood Estimation for estimate the conditional duration. I
am also not sure how to estimate it given various assumption of residual
distribution. It would be a great help for me and I will be very pleased
to have anything related to this issue. Many thanks in advance for your
kind attention.
Attached you will find a zip file with gretl functions for estimating
basic ACD models (exponential, Weibull, Burr) and an example file. The
examples should be more or less self-explanatory for people with some
gretl experience, but please contact me if you need help.
The functions included do only the most basic of jobs, that is
ML estimation with numerical derivatives. Of course there's lots of room
for improvement (analytical derivatives, diagnostics, pretty pictures...).
If you like coding, the functions you'll find in the zipfile should get
you started (hint, hint).
PS. A more up-to-date version of the work linked by Sven appeared in 2008
on the Journal of Economic Surveys.
Riccardo (Jack) Lucchetti
Dipartimento di Economia
Università Politecnica delle Marche
r.lucchetti(a)univpm.it
http://www.econ.univpm.it/lucchetti