Estimating Tukey depth using incremental quantile estimators

作者：

Highlights：

• An approach based in incremental quantile estimators to estimate and track Tukey depth contours is presented.

• The approach is highly memory and computationally efficient.

• The method is demonstrated to detect virtually any changes in multi-variate distributional patterns.

摘要

•An approach based in incremental quantile estimators to estimate and track Tukey depth contours is presented.•The approach is highly memory and computationally efficient.•The method is demonstrated to detect virtually any changes in multi-variate distributional patterns.

论文关键词：Data stream,Incremental quantile estimator,Distributional patterns,Real-time analytics,Tukey depth

论文评审过程：Received 26 August 2020, Revised 13 September 2021, Accepted 18 September 2021, Available online 20 September 2021, Version of Record 25 September 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.108339