| Article ID: | iaor2009636 |
| Country: | Germany |
| Volume: | 3 |
| Issue: | 2 |
| Start Page Number: | 123 |
| End Page Number: | 138 |
| Publication Date: | Jan 1994 |
| Journal: | Central European Journal of Operations Research |
| Authors: | Zhang Jianzhong, Xu Xiaojie, Zhu Detong |
| Keywords: | heuristics |
A nonmonotonic trust region method for unconstrained optimization is proposed. Global convergence and local superlinear rate of the method can be established under general conditions which enable solving the subproblems by simple approximate methods such as two well-known dogleg methods. A numerical comparison with monotonic method is presented which shows some advantages of this revision.