In February 2019, Zhao Junlong, a teacher of our school, published a paper named "Multiple influential point detection in high dimensional regression spaces" in the Journal of the Royal Statistical Society: Series B.

    Due to the masking effect and swamping effect, the diagnosis of the point of influence (abnormal point) in the linear regression of high-dimensional data is a difficult problem. The author defines Max and Min statistics to overcome the masking and swamping effects respectively, and proposes a simple and effective algorithm. Theoretical studies and numerical analysis have proved the effectiveness of the new method.

    The Journal of the Royal Statistical Society is a journal of the Royal Statistical Society. It was established in 1938 and is a quarterly journal. There are only about 8 articles in each issue. It is one of the top four journals recognized in the field of statistics.

　(See the article: https://rss.onlinelibrary.wiley.com/doi/10.1111/rssb.12311)