st: RE: Dirifit / ddirifit: "maximum number of iterations exceeded"

["Nick Cox" <n.j.cox@durham.ac.uk>]

In addition to Maarten's comments I would add that the Dirichlet is a
kind of null model that doesn't include any kind of covariance structure
beyond that implied by constraints on the values or their sum. So, I'd
expect it to perform poorly where there is collaboration or competition
between shares, which seems likely here. 

Nick 
n.j.cox@durham.ac.uk 

Martin Gelter

my coauthor and I are trying to use dirifit to analyze patterns of 
cross-citations between different courts. In order to do so, we have 
transformed our dataset into a matrix showing what percentage of each 
courts' citations points to a particular other court.

Depending on which control variables we use, we sometimes do not get 
results at all (dirifit reports non-converging iterations until the 
user stops it).
Sometimes, when we do get a result with dirifit, ddirifit will result 
in an error message saying "maximum number of iterations exceeded" (I 
typically set the maximum number up to 10000). We suspect that one of 
the problems could be that we have many values in the matrix that are 
very close to zero.

We would be grateful for any suggestions that help us to understand 
the reasons for these problems and how to best address them.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/