An Evaluation of the REMI Model for the South Coast Air Quality Management District

Abstract

This paper reports an evaluation of the econometric model developed by Regional Econometric Models Inc. for the South Coast Air Quality Management District. The analysis is focused on the in-sample performance, forecasting ability, and characteristics in impact analysis of the model. The trade-offs implicit in the performance of this model relate directly to questions of explanatory power. In particular, the model is characterized by well-specified structural equations that enhance its ability to formulate policy-relevant simulations. This may come at the cost of predictive ability in a statistical sense. Choices related to this trade-off are at the heart of applied regional analysis.