A pivoting algorithm solving linear complementarity problems

A matlab implementation of pivoting algorithm solving linear complementarity problems

LCPSolve(M,q) solves the linear complementarity problem:

w = M*z + q, w and z >= 0, w'*z = 0

The function takes the matrix M and the vector q as arguments. The function has three return variables. The first the vectors w and the second is the vector z, found by complementary pivoting. The third return is a 1 by 2 vector. The first component is a 1 if the algorithm was successful and a 2 if a ray termination resulted. The second component is the number of iterations performed in the outer loop.

We acknowledge the work LCPSolve.py in the OpenOpt python package by Rob Dittmar, Enzo Michelangeli and IT Vision Ltd