Fuzzy logic-based portfolio selection with particle filtering and anomaly detection

作者：

Highlights：

• We propose a KBS featuring a fuzzy logic (FL) to create a high-performing portfolio.

• Particle filtering with anomaly detectors is applied to various time-series models.

• Multiple statistical estimations generate a variety of portfolio candidates.

• Our FL system evaluates past records of various portfolios and selects the best one.

• Our FL system integrates multiple investment performance criteria.

摘要

•We propose a KBS featuring a fuzzy logic (FL) to create a high-performing portfolio.•Particle filtering with anomaly detectors is applied to various time-series models.•Multiple statistical estimations generate a variety of portfolio candidates.•Our FL system evaluates past records of various portfolios and selects the best one.•Our FL system integrates multiple investment performance criteria.

论文关键词：Knowledge-based system,Expert system,Fuzzy logic,Investment portfolio,Particle filtering,Anomaly detection

论文评审过程：Received 9 March 2017, Revised 19 May 2017, Accepted 3 June 2017, Available online 6 June 2017, Version of Record 20 June 2017.

论文官网地址：https://doi.org/10.1016/j.knosys.2017.06.006