Regular biography
William Dunsmuir is a professor in the School of Mathematics and Statistics at the University of New South Wales. He is known for his research in time series analysis, particularly in the areas of observation-driven models, Hawkes processes, and their applications in fields such as healthcare, finance, and music perception. Dunsmuir has made significant contributions to the development of statistical methods for analyzing complex data structures, including the glarma package for observation-driven time series regression of counts. His work often bridges theoretical statistics with practical applications, and he has published extensively in top-tier journals such as the Journal of Time Series Analysis, Journal of Applied Probability, and Journal of Statistical Planning and Inference. Dunsmuir has also been involved in interdisciplinary research, including studies on clinical workflow, interruptions in healthcare settings, and the analysis of real-time music perception. His research has been supported by various grants and has had a substantial impact on both academic and applied domains.
Scholar-generated biography
William Dunsmuir is an Emeritus Professor of Statistics at the University of New South Wales, specializing in time series and longitudinal data analysis. His research focuses on statistical modeling of dynamic data, including vector linear time series models, observation-driven models for Poisson counts, and methods for handling missing data in time series. Dunsmuir has also explored applications in clinical settings, such as the impact of interruptions on clinical task completion and medication errors. His work emphasizes robust statistical techniques for analyzing complex data structures, with contributions to both theoretical and applied statistics.