When NLP is used to estimate parameters of a statistical model(using least squares, max Chi-sq or Max likelihood), goodness-of-fit and estimation errors are derived from the inverse of the matrix of second derivatives (Hessian) at the solution point.
Many algorithms produce an estimate of the Hessian as part of their progress, and in other cases it is possible to estimate the matrix in a way similar to estimating the gradient when analytic derivatives are not available. It would be useful to add a function to the nlopt library to extract the Hessian (or its inverse) at the solution. Thanks, Zvi Tarem _______________________________________________ NLopt-discuss mailing list [email protected] http://ab-initio.mit.edu/cgi-bin/mailman/listinfo/nlopt-discuss
