Carlos,
There are likely several problems with your likelihood. You should check it
carefully first before you do any optimization. It seems to me that you have
box constraints on the parameters. They way you are enforcing them is not
correct. I would prefer to use an optimization algorithm t
chine$double.xmax, say that number *1e-6 to avoid
computation troubles) often works, but it is a quick and dirty fix.
JN
On 13-08-13 06:00 AM, r-help-requ...@r-project.org wrote:
Message: 36
Date: Tue, 13 Aug 2013 10:38:05 +0200
From: Carlos Nasher
To:r-help@r-project.org
Subject: [R] [optim/bbmle] fun
Dear R helpers,
I try to find the model parameters using mle2 (bbmle package). As I try to
optimize the likelihood function the following error message occurs:
Error in grad.default(objectivefunction, coef) :
function returns NA at
1e-040.001013016911639890.0003166929388711890.00093516359482939
3 matches
Mail list logo
