Spatial Aggregation of STAR(1,1) Models

• Activity Number: 607
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Business and Economic Statistics Section
• Author(s): Andrew Gehman*+ and William W.S. Wei
• Companies: and Temple University
• Keywords: spatial ; aggregation ; time series ; STARMA ; poolability

Abstract

In this paper, we demonstrate that in most cases the spatial aggregate of spatial time series data contains similar properties to the non-aggregate. In particular, we examine the space-time autoregressive model of temporal and spatial lag 1 (the STAR(1,1) model) and through a simulation study see that even when the condition of poolability does not hold, the aggregate can usually be modeled as a STAR(1,1) model with only slightly higher forecast error. The parameters of the aggregate STAR(1,1) model are also derived and shown to be functions of the non-aggregate parameters. Future work will be to generalize these results to other types of STARMA models.