Review on outlier-tolerant data processing with applications

Owing to the complexity of sampling and running environment, abnormal data such as outliers as well as patchy outliers widely appear in a variety of data from engineering to economic fields. These abnormal data have remarkable bad impact on parameter statistics, system identification, state monitoring, process control, machine learning, decision analysis, and so on. In order to avoid bad impact from abnormal data, a new idea of outlier-tolerant computation was put forward in just the past two decades. In this paper, a brief review is given to describe some major progress and prominent approaches in these fields including outlier-tolerant parameters estimation, outlier-tolerant identification, outlier-tolerant filtering and outlier-tolerant prediction etc. At the end of this paper, several open problems are pointed out for further research.