Consider two random vectors x˜=Az+C11∕2x∈Rp and y˜=Bz+C21∕2y∈Rq, where x∈Rp, y∈Rq z∈Rr are independent with i.i.d. entries of zero mean unit variance, C1 C2 p×p q×q deterministic population covariance matrices, A B p×r q×r factor loading matrices. With n observations x˜ y˜, we study the sample canonical correlations between them. Under sharp fourth moment condition on x, y z, prove BBP transiti...

Feature screening approaches are effective in selecting active features from data with ultrahigh dimensionality and increasing complexity; however, the majority of existing feature either restricted to a univariate response or rely on some distribution model assumptions. In this article, we propose novel sure independence approach based multivariate rank distance correlation (MrDc-SIS). The MrD...

In multivariate linear regression, it is often assumed that the response matrix is intrinsically of lower rank. This could be because of the correlation structure among the prediction variables or the coefficient matrix being lower rank. To accommodate both, we propose a reduced rank ridge regression for multivariate linear regression. Specifically, we combine the ridge penalty with the reduced...

Based on the signal-noise subspace theory, a method of low-rank channel estimation for Orthogonal Frequency Division Multiplexing (OFDM) is presented in this paper. The theory of low-rank linear estimation based on Singular Value Decomposition (SVD) in minimum meansquared error (LMMSE) estimators has been analyzed. A new rank-choose criterion for low-rank approximation is raised. The algorithm’...