A deep dive into multivariate time series at Xiamen University

Renowned globally for his expertise in time series analysis, Professor Li Wai-keung, Dean of the Faculty of Liberal Arts and Social Sciences, continues to share his insights through frequent international lectures. Most recently, he visited Xiamen in Fujian Province to deliver a keynote speech titled “Testing for the Symmetry of Innovation in Multivariate Time Series Models” at the 2025 Symposium on Modern Statistics (2025年廈門大學現代統計學研討會), hosted by Xiamen University. Professor Li gave the first keynote speech at the two-day symposium held on 13 and 14 December 2025.

In his keynote on 13 December, Professor Li highlighted the importance of determining whether innovation in multivariate time series models is symmetrically distributed. He explained, “Understanding this symmetry is essential for both theoretical research and empirical applications. Traditional testing methods for univariate models, which rely on smoothing or martingale transformation treatments, do not extend easily to multivariate contexts.” To address this challenge, he introduced a new consistent test designed to assess diagonal symmetry using a characteristic measure.

Professor Fang Ying, member of the Standing Committee of CCP at Xiamen University and Vice President of the University, delivered the opening speech at the symposium held at the Economics Building of Xiamen University. The event brought together more than 130 scholars from over 40 higher institutions across Chinese Mainland, Hong Kong, Macao, and the United States, and featured seven keynote speeches and 20 parallel sessions. In total, 82 papers were presented, centred on the theme “Frontier Research in Statistics and Data Science in the AI Era”.

It was jointly organised by the School of Economics at Xiamen University, the Wang Yanan Institute for Studies in Economics (WISE, 王亞南經濟研究院), and the Paula and Gregory Chow Institute for Studies in Economics (鄒至莊經濟研究院). The event celebrated the 20th anniversary of the founding of WISE and provided a platform for advancing research in time series analysis, statistical modelling, and data analysis.

Throughout his career, Professor Li has authored more than 176 refereed journal articles and a monograph published by Chapman & Hall. His work is widely cited in scholarly articles, textbooks, and computer software. He is also honoured as a Fellow of both the American Statistical Association and the Institute of Mathematical Statistics.