Nonsmooth DC programming approach to clusterwise linear regression: optimality conditions and algorithms |
| |
Authors: | AM Bagirov J Ugon |
| |
Affiliation: | Faculty of Science and Technology, Federation University Australia, University Drive, Mount Helen, Ballarat, Australia |
| |
Abstract: | The clusterwise linear regression problem is formulated as a nonsmooth nonconvex optimization problem using the squared regression error function. The objective function in this problem is represented as a difference of convex functions. Optimality conditions are derived, and an algorithm is designed based on such a representation. An incremental approach is proposed to generate starting solutions. The algorithm is tested on small to large data sets. |
| |
Keywords: | nonsmooth optimization DC programming regression analysis cluster analysis |
|
|