Name: Liu Xin
Title: Associate Professor
Research Directions: 1. Theoretical issues related to modeling and statistical inference in machine learning, such as parameter testing for support vector machines, random trees, and random forests, classification problems with mislabeled data in training sets, as well as variable selection issues in machine learning and related statistical theory research; 2. Modeling and variable selection issues for (ultra) high-dimensional data, such as time change point selection based on ultra-high-dimensional screening; 3. Joint data modeling methods and theories with temporal and spatial attributes, such as joint modeling of longitudinal data and survival analysis data, directed graphs, and undirected graphs.
Course taught: Data Analysis and Visualization, Mathematical Statistics.
E - mail：liu.xin@mail.shufe.edu.cn
Phone: (021) 6590 2491
Research Project
Talent programs or vertical projects hosted or participated in:
1. October 2021 to present: Shanghai Pujiang Talent Program
2. September 2021 to present: Innovative Research Team of the School of Statistics and Management (Professor You Jinhong's team)
3. September 2020 to present: Youth Innovative Research Team of the School of Statistics and Management (Professor Cui Xiangyu's team)
Research Field
Theoretical research in machine learning, such as statistical theory studies of support vector machines, random trees, and random forests; (ultra) high-dimensional data analysis and modeling, such as ultra high-dimensional variable screening and time change point detection; analysis and joint modeling of longitudinal and time-to-event data; spatial-temporal statistical analysis, such as directed graphs and undirected graphs; functional data analysis; deep learning based on graph data, etc. The types of data studied include complex data, functional data, longitudinal data, and survival data, data with spatiotemporal correlations, image data, etc.; application fields include environmental science, financial industry, medical image analysis (such as MRI and CT images), biomedicine, etc.
Educational Experience
2013-2017 Western University, Canada PhD in Statistics
2012-2013 Canada Western University Master of Statistics
2010-2012 Shanghai University of Finance and Economics Master (School of Statistics and Management, Quantitative Finance and Risk Management Major)
2006-2010 Shanghai University of Finance and Economics Bachelor of Science (School of Statistics and Management, Mathematical Statistics Major)
Work Experience
July 2021 - Present, Shanghai University of Finance and Economics, School of Statistics and Data Science, Associate Professor
February 2019 - June 2021, Shanghai University of Finance and Economics, School of Statistics and Management, Assistant Researcher
February 2018 - January 2019, University of Waterloo and Simon Fraser University, Visiting Scholar
Published or Accepted Papers:
1. Guo, S., Xu, M. & Liu, X.* (2022+). Ultra-high dimensional change point detection. Journal of Multivariate Analysis. Accepted.
2. Liu, X., Zheng, Q., Shen, X., & Wang, S. 2022. An iterative learning algorithm to learn from positive and unlabeled examples. Statistica Sinica, 32, 1-22. https://doi.org/10.5705/ss.202020.0287
3. Liu, X. & He, W. 2022. Adaptive kernel scaling support vector machine with application to a prostate cancer image study. Journal of Applied Statistics, 6(49), 1465-1484. https://doi.org/10.1080/02664763.2020.1870669
4. Zhao, B., Liu, X*., He, W. & Yi, G.Y. 2021. Dynamic tilted current correlation for high dimensional variable screening. Journal of Multivariate Analysis, 182, 104693. https://doi.org/10.1016/j.jmva.2020.104693
5. Liu, X., Yi, G.Y., Bauman, G., & He, W. 2021. Ensembling imbalanced-spatial-structured support vector machine. Econometrics and Statistics, 17, 145-155. https://doi.org/10.1016/j.ecosta.2020.02.003.
6. Fang, Z., Li, W., Liu, X., Pu, X. & Xiang, D. (2021+). Online monitoring of high-dimensional binary data streams with application to extreme weather surveillance, Journal of Applied Statistics. To appear. https://doi.org/10.1080/02664763.2021.1971633
7. Shao, J., Liu, X.* & He, W. 2021. Kernel Based Data-Adaptive Support Vector Machines for Multi-Class Classification. Mathematics, 9(9), 936. https://doi.org/10.3390/math9090936
8. Liu, X., Zhao, B., & He, W. 2020. Simultaneous feature selection and classification for data-adaptive kernel-penalized SVM, Mathematics, 8(10), 1846. https://doi.org/10.3390/math8101846
9. Liu, X., Wu, J., Yang, C. & Jiang, W. 2018. A maximal tail dependence-based clustering procedure for financial time series and its applications in portfolio selection. Risks, 6(115),1-18.
10. Yang, C., Liu, X., Wu, J., Li, Z & Jiang, W. 2018. Clustering of financial time series using jump tail dependence coefficient. Statistical Methods and Applications, 27(3), 491-513.
Working Paper：
1. Zhang, Z. & Liu, X. A novel deep support vector clustering algorithm for unsupervised and semi-supervised learning. Submitted to NeurIPS 2022. Under Review.
2. Wang, S., Shi, H. & Liu, X. Simultaneous dimension deduction and drediction using networks. Submitted to NeurIPS 2022. Under Review.
3. Zhang, Z., Chen, S. & Liu, X. A Novel update and propagation-based dynamic graph neural network with application to consumer finance data. Submitted to ACMKDD 2022. Under Review.
4. Che, Y., Chen, S. & Liu, Xin. Sparse index tracking portfolio with sector neutrality.
Submitted to Electronic Journal of Statistics. Under review.
5. Chen, S., Liu, X. & He, W. Grouped variable selection in joint modeling using block coordinate gradient descent method. Submitted to Canadian Journal of Statistics. Under review.
6. Li, W., & Liu, X. Simultaneous variable selection and covariance estimation in high dimensional linear mixed model. Submitted to Canadian Journal of Statistics. Under review.
