Testing and incorporating seasonal structures into demand models for fruit

Abstract

AbstractIt is widely recognized that purchases of perishable agricultural products are affected by the seasonal cycles of production. When there are seasonal effects where seasonal buying is not explained by prices alone, the seasonal component can be captured using appropriate dummy variables in a demand model. However, over time the introduction of different varieties and foreign sources of supply in the market may affect the seasonal structure in unknown ways, and because of this, it becomes important to identify and test for characteristics of seasonal structure in demand models. In this article, we set out to measure and test the various characteristics of the seasonal component of fruit demand. We model seasonality in the context of an Almost Ideal Demand System (AIDS) model, where we consider several types of fruit demand in the United States.