Correlation Analysis of Long-Term Financial Time Series by a Fuzzy Trend Model

N. Watanabe and M. Kuwabara (Japan)

Keywords

Trend, TakagiSugeno’s model, Fuzzy IfThen Rules, Cross correlation, Stock return

Abstract

A method for analyzing multivariate financial time series such as stock returns is considered by introducing a fuzzy trend model. A fuzzy trend model is based on fuzzy if then rules and suitable for financial time series. The fuzzy trend model proposed in this note is for multivariate time series. This model provides a correlation analysis method for trends of long-term series. Such analysis cannot be achieved by usual models. Applicability of the proposed model is shown by applying to real time series data.

Important Links:



Go Back