A0150
Title: Multivariate spatiotemporal models with low rank coefficient matrix
Authors: Qingzhao Zhang - Xiamen University (China) [presenting]
Abstract: Multivariate spatiotemporal data arise frequently in practical applications, often involving complex dependencies across cross-sectional units, time points and multivariate variables. In the literature, few studies jointly model the dependence in three dimensions. A multivariate reduced-rank spatiotemporal model is proposed to model the cross-sectional, dynamic, and cross-variable dependence simultaneously. By imposing the low-rank assumption on the spatial influence matrix, the proposed model achieves substantial dimension reduction and has a nice interpretation, especially for financial data. Due to the innate endogeneity, the quasi-maximum likelihood estimator (QMLE) is proposed to estimate the unknown parameters. A ridge-type ratio estimator is also developed to determine the rank of the spatial influence matrix. The asymptotic distribution of the QMLE and the rank selection consistency of the ridge-type ratio estimator is established. The proposed methodology is further illustrated via extensive simulation studies and two applications to stock market and air pollution datasets.
