Home Article

Functional data analysis with covariate-dependent mean and covariance structures


Speaker: LIN Huazhen,Southwestern University of Finance and Economics

Venue: Online,Tencent Conference, ID: 441 837 276

Abstract: Functional data analysis has emerged as a powerful tool in response to the ever increasing  resources and efforts devoted to collecting information about response curves or anything varying over a continuum. However, limited progress has been made to link the covariance structure of response curves to external covariates, as most functional models assume a common covariance structure. We propose a new functional regression model with covariate-dependent mean and covariance structures. Particularly, by allowing the variances of the random scores to be covariate-dependent, we identify eigenfunctions for each individual from the set of eigenfunctions which govern the patterns of variation across all individuals, resulting in high interpretability and prediction power. We further propose a new penalized quasi-likelihood procedure, which combines regularization and B-spline smoothing, for model selection and estimation, and establish the convergence rate and asymptotic normality for the proposed estimators. The utility of the method is demonstrated via simulations as well as an analysis of the Avon Longitudinal Study of Parents and Children on parental effects on the growth curves of their offspring, which yields biologically interesting results.