Gaussian process methods for nonparametric functional regression

Speaker:  Dr Bo Wang, Leicester University

Abstract

Functional data analysis (FDA) has received increasing interests in recent years, and a large number of statistical methods have been developed and applied in a wide range of fields such as finance, medicine, chemical engineering, biology, etc. In this talk I'll introduce Gaussian process methods for nonparametric functional regression for both scalar and functional responses with mixed multidimensional functional and scalar predictors. The proposed models allow the response variables depending on the entire trajectories of the functional predictors, and provide a general framework to incorporate both scalar and functional predictors of high dimension. They inherit the desirable properties of Gaussian process regression, and the uncertainty in predictions can also be easily obtained and expressed. The numerical experiments show that the proposed methods significantly outperform the competing models, and their usefulness is also demonstrated by the application to two real datasets.