[Octave-bug-tracker] [bug #41120] QP function result is not correct

[Mariotto Mauro]

URL:
  <http://savannah.gnu.org/bugs/?41120>

                 Summary: QP function result is not correct
                 Project: GNU Octave
            Submitted by: mariottomauro
            Submitted on: Mo 06 Jan 2014 10:02:08 GMT
                Category: Octave Function
                Severity: 3 - Normal
                Priority: 5 - Normal
              Item Group: Incorrect Result
                  Status: None
             Assigned to: None
         Originator Name: 
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any
                 Release: 3.6.2
        Operating System: GNU/Linux

    _______________________________________________________

Details:

Consider the following quadratric problem.

min f(x) = 0.5*x'Hx + Q'x

and the following example

min -3x1^2 -9x1x2 -3.5x2^2

sub 2x1 + 3x2 <= 12
x1>=0, x2>=0

Define the following matrices and vectors for the above problem:
A=[]
B=[]
Q=[]
LB=[0,0]'
A_IN=[2,3]
A_UB=[12]
H=[-6,-9; -9,-7]

Call the build-in Octave function

[X,F,I,L]=qp([],H,Q,A,B,LB,[],[],A_IN,A_UB)

Incorrect result given by Octave 3.6.2 and 3.6.4 in Ubuntu Linux 12.10 and
13.10 i686

x1=0
x2=0
f(x1,x2)=0

The correct result is however:

x1=6
x2=0
y1=18 Lagrange Multiplier or KKT Multiplier
f(x1,x2)=-108





 





    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?41120>

_______________________________________________
  Nachricht gesendet von/durch Savannah
  http://savannah.gnu.org/