Research

Publications

S. Shen, H. Zhou, K. He, L. Zhou. Principal Component Analysis of Two-dimensional Functional Data with Serial Correlation. The Journal of Agricultural, Biological, and Environmental Statistics, in press. (https://doi.org/10.1007/s13253-023-00585-8) 

S. Shen, K. He, B. Zhang, L. Zhou. Exponential-family Principal Component Analysis of Two-dimensional Functional Data with Serial Correlation. Submitted.

Q. Xia, H. Wong, S. Shen, K. He (2022). Factor Analysis for High-dimensional Time Series: Consistent Estimation and Efficient Computation. Statistical Analysis & Data Mining. 15 (2022), 247-263. (https://doi.org/10.1002/sam.11557)

Research Interests

Digital Earth and Smart City

Digital Earth is an interdisciplinary research field associated with geography science, remote sensing, internet of things, computer and data science. It aims to develop a visualizable, and intelligent digital platform of our planet from all possible perspectives. The large-scale data is the key issue in this field and we hope to find the faster and more accurate models for many industrial applications, like smart city, transportation research, environmental science, sustainable development.

Spatio-Temporal Data Mining

Data with location and time attributes are ubiquitous. How to obtain the spatiotemporal patterns from the data is a challenging problem in data science. This topic includes exploring the spatiotemporal clustering, correlation, and regression. It will be applied to analyze the data from multiple scenarios like Internet of Things (IoTs), environmental science, and location based services (LBS).

Functional Data Analysis

Functional data analysis (FDA) is a branch of statistics that analyzes curves, surfaces, sound waves or anything that can be considered as functions. The basic statistical methods cannot work well for analyzing functional data, so we need to develop new methods to handle this novel structure of data. The methods we proposed in FDA can deal with some new real world problem, which mean that there are broad prospects for development in FDA.

Dynamic Network Analysis

Another research topic of mine is dynamic network analysis. As we know, networks (graphs) are gradually becoming one of the main data structures in our daily life, including social networks, economical networks, gene regulatory networks and so on. Researchers often focused on the static networks analysis, which lost the temporal information contained in the dynamic network structures. We aim to discover the dynamic patterns in networks and help decision makers making great strategies.