Model Selection and Forecasting Ability of Theory-Constrained Food Demand Systems

TLDR: In this article, the authors used out-of-sample forecasting of annual U.S. per capita food consumption, applying data from 1923 to 1992, is used as a basis for model selection among the absolute price Rotterdam model, a first-differenced linear approximate almost ideal demand system (FDLA/ALIDS) model, and a double-log demand system.

Abstract: Out-of-sample forecasting of annual U.S. per capita food consumption, applying data from 1923 to 1992, is used as a basis for model selection among the absolute price Rotterdam model, a first-differenced linear approximate almost ideal demand system (FDLA/ALIDS) model, and a first-differenced double-log demand system. Conditional-on-price consumption forecasts derived from elasticities are determined to be superior to direct statistical model forecasts. Models with consumer theory imposed through parametric restrictions provide better forecasts than models with little theory-imposition. For these data, a double-log demand system is a superior forecaster to the Rotterdam model, which is superior to the FDLA/ALIDS model.