A0740
Title: Asymptotic theory in spatiotemporal unstable autoregression
Authors: Zhishui Hu - University of Science and Technology of China (China)
Hanying Liang - Tongji University (China)
Zudi Lu - City University of Hong Kong (China)
Qiying Wang - University of Sydney (Australia)
Zudi Lu - City University of Hong Kong (China) [presenting]
Abstract: The issue of estimation and unit root test has been essential and critical for the analysis of time series, but it remains less addressed for spatial time series data. The estimation and unit root test are first extended from a time series to spatiotemporal unstable AR(1) model. A local location coefficient fitting is proposed to overcome the location-wide heteroscedasticity, and the estimation procedure and unit root test are developed with the asymptotic properties established. Further extension to the spatiotemporal AR(p) model with unit root test is developed. The theory and method can be seen as an extension of the popular augmented Dickey-Fuller (ADF) test in time series analysis to the spatiotemporal data setting. Both asymptotic theory and simulations clearly demonstrate that the spatiotemporal estimation and unit root test are more efficient and reliable than the time series only based estimation procedure and traditional unit root test.
