[R] Nonlinear programming problem
Ravi Varadhan
rvaradhan at jhmi.edu
Fri Sep 17 19:55:33 CEST 2010
Hi,
You have not provided a reproducible example, so I can only make an educated
guess. This is likely a numerical problem with trying to solve a linear
system of equations in the Lagrange-Newton method (sequential quadratic
programming or SQP) for solving nonlinearly programming problems. I am not
sure what you can do about this, but here is a strategy:
1. be try a different, more reasonable, starting value.
2. If that does not help, you may want to check your problem specification
to make sure that everything (objective function, constraints) is specified
correctly
3. If (1) and (2) do not help, look at the scaling of the parameters,
objective function, and the scaling of equality and inequality constraints.
You can sometimes get convergence by simple scaling so that all these
different entities are roughly of the same order of magnitude.
4. If all else fails, try a different algorithm. For example, the "alabama"
package has `constrOptim.nl'
function that you could try.
But in future, when you ask these questions you should always send us a
minimally self-contained, reproducible example, so that we can better help
you.
Hope this is helpful,
Ravi.
-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Xiaoxi Gao
Sent: Friday, September 17, 2010 2:03 AM
To: R Help
Subject: [R] Nonlinear programming problem
Hello useRs,
I'm using the command "solnp" in package "Rsolnp" to solve a general
nonlinear programming problem. But I got an error that " the leading minor
of order 15 is not positive definite
". Can anybody tell what may cause this error? Does it have something to do
with the starting values?
Thanks a lot!!!
Xiaoxi
[[alternative HTML version deleted]]
______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
More information about the R-help
mailing list