Temporal and Spatial Aggregation: Alternative Marketing Margin Models

Abstract

AbstractThe effects of temporal and spatial data aggregation on the performance of alternative marketing margin models is assessed using monthly, quarterly, and semiannual fluid milk prices from three U.S. cities. Nonnested tests for multivariate and single equation models with serial correlation are used to choose among alternative models at each aggregation level. While model choice is affected by temporal and spatial aggregation, model choice becomes more difficult as data are temporally or spatially aggregated.