A0165
Title: Scalable semiparametric spatio-temporal regression for large data analysis
Authors: Jun Zhu - University of Wisconsin - Madison (United States) [presenting]
Abstract: With the rapid advances in data acquisition techniques, spatio-temporal data are becoming increasingly abundant in a diverse array of disciplines. Spatio-temporal regression methodology is developed for analyzing large amounts of spatially referenced data collected over time, motivated by environmental studies utilizing remotely sensed satellite data. In particular, a semiparametric autoregressive model is specified without the usual Gaussian assumption and a computationally scalable procedure is devised that enables the regression analysis of large datasets. The model parameters are estimated by maximum pseudo-likelihood, and the computational complexity can be reduced from cubic to linear of the sample size. Asymptotic properties under suitable regularity conditions are further established that inform the computational procedure to be efficient and scalable. A simulation study is conducted to evaluate the finite-sample properties of the parameter estimation and statistical inference. The methodology is illustrated by a dataset with over 2.96 million observations of annual land surface temperature, and a comparison with an existing state-of-the-art approach to spatio-temporal regression highlights the advantages of our method.
