Rate My Professor Mengyu Xu

Xu's research interests encompass covariance matrix estimation and time-varying network recovery from high-dimensional time series, the distribution theory of quadratic forms and high-dimensional hypothesis testing, time series analysis, high-dimensional models, high-dimensional testing and inference, non-linear processes, and change point detection. Her key publications include "Covariance and precision matrix estimation for high-dimensional time series" (Annals of Statistics, 2013, with Xiaohui Chen and Wei Biao Wu), "Regularized estimation of linear functionals of precision matrices for high-dimensional time series" (IEEE Transactions on Signal Processing, 2016, with Xiaohui Chen and Wei Biao Wu), "Pearson’s chi-squared statistics: approximation theory and beyond" (Biometrika, 2019, with Danna Zhang and Wei Biao Wu), "High accuracy machine learning identification of fentanyl-relevant molecular compound classification via constituent functional group analysis" (Scientific Reports, 2020, with Chun-Hung Wang, Anthony C. Terracciano, Artem E. Masunov, and Subith S. Vasu), and "Good and bad self-excitation: Asymmetric self-exciting jumps in bitcoin returns" (Economic Modelling, 2023, with Chuanhai Zhang, Zhengjun Zhang, and Zhe Peng). As principal investigator, she has obtained grants from Northrop Grumman Corporation, United Negro College Fund, Casualty Actuaries of the Southeast, Siemens Inc., and UCF seed funding programs. Xu received the Best Paper Award at the ASME Turbo Expo for "Prediction enhancement of machine learning using time series modeling in gas turbines" in 2021. She has chaired Ph.D. dissertations, served on 19 Ph.D. committees, mentored master's and undergraduate students, organized sessions at the 2022 Joint Statistical Meetings and the 2019 ASA Florida Chapter Meeting, and reviewed manuscripts for journals including Annals of Statistics, Bernoulli, and Journal of Multivariate Analysis.