EcoSta 2018: Registration
View Submission - EcoSta2018
Title: Spatio-temporal autoregressions with network information Authors:  Yingying Ma - Beihang University (China) [presenting]
Shaojun Guo - Renmin University of China (China)
Qiwei Yao - London School of Economics (UK)
Abstract: A new class of spatio-temporal models with unknown autoregressive coefficient matrices is proposed. The setting represents a sparse structure for high-dimensional spatial panel dynamic models when panel members represent economic (or other type) individuals at many different locations. Due to the innate endogeneity, we apply two different approaches to estimate the autoregressive coefficient matrices. A variable selection method has been developed for determining the unknown elements in the autoregressive matrices. Some asymptotic properties of the inference methods are established. The proposed methodology is further illustrated using both simulated and real data sets.