Spatial Autoregressive Conditional Heteroskedasticity Models

Abstract

This study proposes a spatial extension of time series autoregressive conditional heteroskedasticity (ARCH) models to those for areal data. We call the spatially extended ARCH models as spatial ARCH (S-ARCH) models. S-ARCH models specify conditional variances given surrounding observations, which constitutes a good contrast with time series ARCH models that specify conditional variances given past observations. We estimate the parameters of S-ARCH models by a two-step procedure of least squares and the quasi maximum likelihood estimation, which are validated to be consistent and asymptotically normal. We demonstrate the empirical properties by simulation studies and real data analysis of land price data in Tokyo areas.