Module MNH in PORT

• PORT Information

• Components
  • Fullsource from NETLIB
  • Source from NETLIB

MNH

 
Finds a local minimum of a continuously differentiable function. User supplies
gradient and Hessian of objective function. Uses a variant of Newton's method
with a model/trust-region technique to aid convergence from poor starting
values. (This is subroutine HUMSL of ACM Trans. Math. Softw. 9 (1983) 503-524.).
 
Classes  :  G1b1c . Unconstrained optimization of a smooth multivariate
                  function, user provides first and second derivatives
 
Type     : Fortran subroutine in PORT library (NL2OPT sublibrary).
Access   : Proprietary. Many implementations available.
Precision: Single.
 
Usage    : CALL MNH (N, D, X, CALCF, CALCGH, IV, LIV, LV, V, UIPARM, URPARM,
           UFPARM)
 
Details  : Fullsource Source
Sites    : (1) NETLIB
 

Implementation of MNH from PORT on NETLIB

 
NETLIB:    Public access repository, The University of Tennessee at
           Knoxville and Bell Laboratories
 
Precision: Single. (Double: DMNH)
 
You may access components from NETLIB outside GAMS as follows.
 
   Source       : Available by anonymous ftp from netlib in the port
                  subdirectory
   Fullsource   : Available by http from netlib in the port subdirectory