I develop statistical methods for complex time series, functional data, and longitudinal signals, with an emphasis on nonstationary processes, spectral analysis, Bayesian learning, and computationally efficient tools for modern biomedical and scientific applications.
Scott A. Bruce is an Associate Professor of Statistics at Texas A&M University. His work combines methodological and collaborative statistics, with research centered on nonstationary and multivariate time series, adaptive frequency band analysis, Bayesian spectral modeling, and scalable statistical learning for complex data.
Much of his research is motivated by transdisciplinary problems in sleep and circadian science, cardiology, psychiatry, neuroscience, biomechanics, and digital health. Across these settings, he develops practical, computationally efficient methods for extracting interpretable structure from high-dimensional signals, wearable measurements, and longitudinal observations.
His recent work includes frequency band analysis for multiple stationary and nonstationary multivariate time series, spectral methods for categorical and functional time series, Bayesian tree-based models for covariate-dependent spectra, and distributed learning methods for large heterogeneous data.
Current themes reflect the methodological and applied breadth of the research portfolio.
My research develops statistical methodology for the analysis of temporally structured data arising in modern observational and experimental studies. I am especially interested in methods that are interpretable, adaptive, and computationally feasible in real-world scientific collaborations.
Selected software and methodological projects that reflect ongoing work in statistical computing and frequency-domain analysis.
R code for multivariate empirical band analysis, including simulated examples and electroencephalography applications that accompany the frequency band analysis work for multiple time series.
View repository
Software and methods for dimension reduction and adaptive frequency band learning in functional time series, motivated by complex biomedical signals and other high-dimensional temporal data.
View repository
R code for classification and clustering of categorical time series using the spectral envelope and optimal scalings, connecting interpretable frequency-domain summaries with modern machine learning tasks.
View repository
Methods and code for covariate-dependent spectral analysis of multiple time series using flexible Bayesian tree ensembles, with applications to complex biomedical studies.
View repositoryAll publications are shown below in CV order. Use the search box to filter by author, title, journal, year, or keyword.
Recent teaching includes graduate and undergraduate data science and statistics courses at Texas A&M University, along with earlier graduate teaching at George Mason University and undergraduate teaching at Temple University.
Graduate courses
Undergraduate courses
George Mason University
Temple University