Rate My Professor Samuel Kou

Samuel Kou is the Abbott Lawrence Lowell Professor of Statistics in Harvard University’s Faculty of Arts and Sciences and Professor of Biostatistics in the Harvard T.H. Chan School of Public Health. He earned a B.S. in Computational Mathematics from Peking University in 1997, an M.S. in Statistics from Stanford University in 2000, and a Ph.D. in Statistics from Stanford University in 2001. Kou joined the Harvard faculty as Assistant Professor of Statistics in 2001, advanced to John L. Loeb Associate Professor of the Natural Sciences in 2005, became Professor of Statistics in 2008, and served as Chair of the Department of Statistics starting July 2018. His research centers on stochastic inference in single molecule biophysics, chemistry, and biology; Bayesian inference of stochastic models; nonparametric methods, model selection, and empirical Bayes; Monte Carlo methods; and economic and financial modeling. Kou pioneered stochastic modeling of nanoscale biophysics through collaborations with X. Sunney Xie, developing models to explain random phenomena in single-molecule experiments that challenge classical Brownian diffusion theories. He transformed Monte Carlo methods by introducing the equi-energy sampler for efficient simulations in complex systems. In collaboration with Bradley Efron, he provided a theoretical framework for non-parametric regression and model selection with broad applications across fields. Kou also modeled growth stocks, biotechnology, and internet stocks using stochastic processes in work with Steven Kou.

Kou received the 2012 COPSS Presidents’ Award for outstanding statistical contributions, the inaugural Young Investigator Award in 2009, and a Guggenheim Fellowship. Key publications include “Equi-energy sampler with applications in statistical inference and statistical mechanics” (Ann. Statist., 2006), “Bayesian analysis of single-molecule experimental data” (J. Roy. Statist. Soc. C, 2005), “Stochastic modeling in nanoscale biophysics: subdiffusion within proteins” (Ann. Appl. Statist., 2008), “A multiresolution method for parameter estimation of diffusion processes” (J. Amer. Statist. Assoc., 2012), and “SURE estimates for a heteroscedastic hierarchical model” (J. Amer. Statist. Assoc., 2012). His innovations have profoundly influenced interdisciplinary statistics, enhancing analysis in biophysics, public health through internet search data for disease tracking, and computational biology including protein conformation methods.