Low-Rank Covariance Function Estimation for Multidimensional Functional Data

Multidimensional function data arise from many fields nowadays. The covariance function plays an important role in the analysis of such increasingly common data. In this article, we propose a novel nonparametric covariance function estimation approach under the framework of reproducing kernel Hilbert spaces (RKHS) that can handle both sparse and dense functional data. We extend multilinear rank structures for (finite-dimensional) tensors to functions, which allow for flexible modeling of both covariance operators and marginal structures. The proposed framework can guarantee that the resulting estimator is automatically semipositive definite, and can incorporate various spectral regularizations. The trace-norm regularization in particular can promote low ranks for both covariance operator and marginal structures. Despite the lack of a closed form, under mild assumptions, the proposed estimator can achieve unified theoretical results that hold for any relative magnitudes between the sample size and the number of observations per sample field, and the rate of convergence reveals the phase-transition phenomenon from sparse to dense functional data. Based on a new representer theorem, an ADMM algorithm is developed for the trace-norm regularization. The appealing numerical performance of the proposed estimator is demonstrated by a simulation study and the analysis of a dataset from the Argo project.

Automated summary

In this article, a novel nonparametric covariance function estimation approach is proposed for multidimensional function data. The approach utilizes multilinear rank structures and reproducing kernel Hilbert spaces to model covariance operators and marginal structures flexibly. The resulting estimator is automatically semipositive definite and can incorporate various spectral regularizations, with trace-norm regularization promoting low ranks. The proposed approach achieves unified theoretical results for different sample sizes and data densities, revealing a phase-transition phenomenon from sparse to dense data.