Report title: The Marriage between Econometrics and Machine Learning: New Inference for High-Dimensional Sparse Spatio-Temporal Autoregressions

Reporter: Associate Professor Guo Shaojun Renmin University of China

Reporting time: 16:00-17:00, November 20, 2020

Report location: Tianyuan fifth seminar room on the third floor of the Mathematics Building

School contact: Zhu Fukang fzhu@jlu.edu.cn

Report summary:

    We consider a new class of spatio-temporal models with sparse autoregressive coefficient matrices and exogenous variable. To estimate the model, we first profile the exogenous variable out of the response. This leads to a profiled model structure. Next, to overcome endogeneity issue, we propose a class of generalized methods of moment (GMM) estimators to estimate the autoregressive coefficient matrices. A novel bagging-based estimator is further developed to conquer the over-determined issue which also occurs in Chang et al. (2015) and Dou et al. (2016). An adaptive forward-backward greedy algorithm is proposed to learn the sparse structure of the autoregressive coefficient matrices. A new BIC-type selection criterion is further developed to conduct variable selection for GMM estimators. Asymptotic properties are further studied. The proposed methodology is illustrated with extensive simulation studies.

Brief introduction of the speaker:

    Guo Shaojun is an associate professor at the Institute of Statistics and Big Data, Renmin University of China. He graduated from Shandong Normal University with a bachelor's degree in 2003 and received a PhD in Science from the Academy of Mathematics and Systems Science, Chinese Academy of Sciences in 2008. After graduating from his Ph.D., he worked at the Academy of Mathematics and Systems Science, Chinese Academy of Sciences, as an assistant researcher, until 2016. From 2009 to 2010, he went to the Department of Operations Research and Financial Engineering of Princeton University in the United States for post-doctoral research, doing research on high-dimensional data analysis, and from 2014 to 2016 in the London School of Economics, United Kingdom, doing post-doctoral research in the Department of Statistics, London School of Economics, doing large-dimensional time series construction Research on models. The current main research directions are: statistical learning; non-parametric and semi-parametric statistical modeling; survival analysis and functional data analysis.