Viet Anh Nguyen wins the 2018 George Nicholson Award

The George Nicholson Student Paper Competition, arguably the most prestigious student award in the operations research community, is held each year since 1975 to honor outstanding student papers in the field of operations research and the management sciences. Viet Anh is the first winner from a European university. He received the Nicholson Award for his paper “Distributionally Robust Inverse Covariance Estimation: The Wasserstein Shrinkage Estimator.” The paper proposes a principled approach for estimating the precision matrix (i.e., the inverse covariance matrix) of a Gaussian random vector from independent samples by leveraging ideas from modern distributionally robust optimization. The idea is to solve a worst-case maximum likelihood estimation problem, where the worst case is evaluated across all Gaussian distributions within a prescribed Wasserstein distance from a Gaussian reference distribution characterized by the sample mean and the sample covariance matrix. The precision matrix is a key ingredient, for instance, for the optimal classification rule in linear discriminant analysis, the optimal investment portfolio in Markowitz’ celebrated mean-variance model or the optimal array vector of the beamforming problem in signal processing. Moreover, the optimal fingerprint method used to detect a multivariate climate change signal blurred by weather noise requires knowledge of the climate vector’s precision matrix. Viet Anh showed that his new estimator has many desirable theoretical properties and displays a similar performance as the graphical lasso approach (which requires solving a large semidefinite program) at the computational cost of a naive linear shrinkage estimator (which requires merely a spectral decomposition).