Weighted score-driven fuzzy clustering of time series with a financial application

作者：

Highlights：

• The paper proposes a new method for time series clustering.

• The procedure is based on an optimal combination of conditional and unconditional moments.

• The clustering procedure includes fuzziness.

• The clustering method is applied to financial time series.

摘要

•The paper proposes a new method for time series clustering.•The procedure is based on an optimal combination of conditional and unconditional moments.•The clustering procedure includes fuzziness.•The clustering method is applied to financial time series.

论文关键词：Fuzzy clustering,Dynamic Conditional Score,Conditional moments,Unconditional moments,Optimal weighting procedure for clustering

论文评审过程：Received 22 October 2021, Revised 17 February 2022, Accepted 23 February 2022, Available online 5 March 2022, Version of Record 18 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116752