A Fuzzy Model for Long-term Financial Time Series

N. Watanabe and M. Kuwabara (Japan)

Keywords

Fuzzy IfThen Rules, nonlinear model, ARCH model, stock return

Abstract

A time series model based on fuzzy if-then rules is intro duced for long-term financial time series such as stock re turns. Usually the level of time series is assumed to be con stant when the ARCH or GARCH model is fitted to time series. However this assumption does not hold for the long term time series. The proposed model permits the changing level by introducing the latent variables. The applicability of the proposed model is considered by applying to real time series, and the possibility of the detailed economical analysis is discussed by using estimated latent variables.

Important Links:



Go Back