Detection of change points in time series analysis with fuzzy statistics

Changes in time series often occur gradually so that there is a certain amount of fuzziness in the change point. In this paper we have presented an integrated identification procedure for change-point detection based on fuzzy logic. The membership function of each datum corresponding to the cluster centres is calculated and is used for performance index grouping. We have also suggested a test for the change in level and the change in slope for testing a hypothesis about change points. We have made simulation studies to demonstrate the whole procedure. Finally an empirical study about change-point identification in the exchange rate data of six Asian nations has been demonstrated using the algorithm of the paper.