On Mon, 3 Dec 2012, Rezart Hoxhaj wrote:
Dears
My problem is the following. I have data on income in both values and
income intervals. Using a log likelihood for both sub
samples, I want to get parameters of the joint sample. My problem is how to
define the income intervals in Gretl, and apply them to a
cumulative density function, normally distributed ( not standard cdf).
I want to express the income intervals as a difference between
cumulative density functions at each point. Any idea how to do that??
Hope I’m clear. It is clear that I’m a beginner in Gratl.
If I understand correctly what your problem is, your model may be
estimated via the intreg command. If you want to use other distributions
than the normal, then you need to write a script and use the mle command.
-------------------------------------------------------
Riccardo (Jack) Lucchetti
Dipartimento di Scienze Economiche e Sociali (DiSES)
Università Politecnica delle Marche
(formerly known as Università di Ancona)
r.lucchetti(a)univpm.it
http://www2.econ.univpm.it/servizi/hpp/lucchetti
-------------------------------------------------------