Michael W. Trosset

Michael W. Trosset is a Professor of Statistics and Director of Graduate Studies at Indiana University Bloomington's Department of Statistics. With a PhD in Statistics from UC Berkeley (1983) and a BA in Mathematics from Rice University (1978), he brings over 40 years of academic experience to his role. His career includes faculty positions at the University of Arizona (14 years) and the College of William & Mary (8 years) before joining Indiana University.

Professor Trosset's educational background includes:

• B.A. in Mathematics and Mathematical Sciences, Rice University, 1978
• Ph.D. in Statistics, University of California Berkeley, 1983

His research focuses on computational statistics and statistical learning, with particular expertise in multidimensional scaling, manifold learning, and high-dimensional data analysis. He investigates how proximity data can be represented in Euclidean space and how dimension reduction techniques can inform subsequent statistical inference. His work spans theoretical foundations and practical applications in network analysis, particularly random dot product graphs.

Professor Trosset's publication record shows consistent contributions to statistical methodology, with recent work emphasizing the theoretical underpinnings of manifold learning for statistical inference. His research bridges computational mathematics, statistical theory, and practical data analysis applications, particularly in high-dimensional settings where traditional multivariate methods face challenges.

As an educator, Professor Trosset has developed a distinctive approach that emphasizes understanding statistical principles rather than treating methods as black boxes. He is the author of the textbook An Introduction to Statistical Inference and Its Applications with R (2009) and teaches courses ranging from introductory statistics to advanced topics in statistical computing and high-dimensional data analysis.

His teaching philosophy centers on helping students understand why statistical methods work, with emphasis on mathematical notation and computational implementation using R. He encourages students to move beyond superficial coverage of many topics to develop deeper understanding of fewer methods. His courses typically include custom R functions to illustrate statistical concepts and algorithms.

Professor Trosset maintains an active research collaboration with Carey Priebe at Johns Hopkins University, co-teaching the course 'Manifold Learning for Subsequent Inference' and publishing jointly on topics related to statistical inference on network data using dimension reduction techniques.