FUZZY CLUSTERING ANALYSIS OF THE EARLY WARNING SIGNS OF FINANCIAL CRISIS

摘要

The fuzzy logic approach offers us the means to deal with uncertainty inherent in a wide variety of tasks, to treat vague and imprecise linguistic values, and to model nonlinear relationships. In this paper we apply fuzzy clustering methods and tools to the evaluation of early warning indicators of financial crises. We first provide the relevant background domain knowledge and men present the fuzzy clustering approach that we used. We also give empirical results from the fuzzy clustering analysis of the Finnish currency crisis in 1992.