[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: No Convegence in Likelihood Estimation

From   Maarten buis <[email protected]>
To   [email protected]
Subject   Re: st: No Convegence in Likelihood Estimation
Date   Wed, 16 Jan 2008 20:31:43 +0000 (GMT)

--- Mostafa Baladi <[email protected]> wrote:

> Dear Statalist members,
> I am estimating different ARIMA orders for the same data set.  For
> some, the likelihood maximization does not converge and I get the
> following error message:  
> flat log likelihood encountered, cannot find uphill direction
> I reviewed the help options and I am using gtolerance(999) which is
> supposed to disable the gradient criterion when it is difficult to
> achieve convergence.  I still get the same error message.  Can you
> please give me some advice on this?

Often non-convergence is a sign that there is something wrong with your
model. You can set the -trace- option and see if one the parameter
estimates is very very large, or very very small. That would be an
indication of the source of the problem. Also you can experiment by
specifying different maximization methods in the -technique()- option.

-- Maarten

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Yahoo! Answers - Got a question? Someone out there knows the answer. Try it
*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index