-1#ind in mincgoptimize
Page 1 of 1

Author:  philrosenberg [ Wed Feb 18, 2015 9:36 pm ]
Post subject:  -1#ind in mincgoptimize

I am using mincgoptimize to find the maximum of a multidimensional probability distribution function. Unfortunately after a few iterations my callback receives -1#Ind from alglib. I'm not sure exactly what is causing the problems, I wonder if it has to do with the numerical differentiation that is being performed as the gradients are likely to be very shallow in the wings of the distribution.

Does anyone have any idea how I could solve this or how I could debug the issue?


Page 1 of 1 All times are UTC
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group