CIRJE-F-1037 "Fuzzy Logic-based Portfolio Selection with Particle Filtering and Anomaly Detection"
Author Name

Nakano, Masafumi, Akihiko Takahashi and Soichiro Takahashi

Date February 2017
Full Paper  
Remarks  Revised in June 2017; Subsequently published in Knowledge-Based Systems, Volume 131, 1 September 2017, Pages 113–124.

This paper proposes a new knowledge-based system (KBS) featuring fuzzy logic (FL) with particle filtering and anomaly detection to create high-performance investment portfolios. In particular, our FL system selects a portfolio with fine risk-return profiles from a number of candidates by integrating multilateral performance measures. The candidates consist of various portfolios based on multiple time-series models estimated by a particle filter with anomaly detectors. In an out-of-sample numerical experiment with a dataset of international financial assets, we demonstrate our KBS successfully generates a series of selected portfolios with satisfactory investment records.