Fuzzy c-regression models combined with support vector regression

Fuzzy c-regression models (FCRM) give us multiple clusters and regression models of each cluster simultaneously, while support vector regression models (SVRM) involve kernel methods which enable us to analyze non-linear structure of the data. We combine these two concepts and propose the united fuzzy c-support vector regression models (FC-SVRM). In case that c is unknown, we introduce sequential regression models (SRM) into SVRM, and propose support vector sequential regression models (SVSRM). We show numerical examples to compare results from these methods.