Distributed sparse composite quantile regression in ultrahigh dimensions

We study the sparse composite quantile regression under ultrahigh dimensionality and make theoretical and algorithmic contributions.

Density-Convoluted Support Vector Machines for High-Dimensional Classification

We study the density convoluted support vector machine (DCSVM) and provide an efficient algorithm for solving it which can be more than 100 times faster than that for the standard SVM.

Demystifying a class of multiply robust estimators

We demystify the multiply robust estimators under the missing at random assumption and show they are essentially doubly robust estimators.

A Unified Approach to Sparse Tweedie Modeling of Multisource Insurance Claim Data

We study multi-source insurance data using carefully designed group regularization and provide an efficient algorithm to solve the problem.

Aggregated Expectile Regression by Exponential Weighting

We propose a nonparametric approach to expectile regression estimation via model combination.

ADMM for High-Dimensional Sparse Penalized Quantile Regression

We propose two variants of the ADMM algorithm to solve the weighted lasso and elastic net penalized quantile regression and show they are very efficient.

High-Dimensional Generalizations of Asymmetric Least Squares Regression and Their Applications

We consider sparse penalized asymmetric least squares regression and its generalizations.

Pulse-number discrimination by Cope’s gray treefrog (Hyla chrysoscelis) in modulated and unmodulated noise

We study the effect of modulated noise on discrimination among different conspecific calls by treefrogs in a simulated environment