A FRAMEWORK FOR REGIONAL MODELING AND IMPACT ANALYSIS: AN ANALYSIS OF THE DEMAND FOR ELECTRICITY BY LARGE MUNICIPALITIES IN ONTARIO,CANADA*

ABSTRACT. We propose a mixed fixed and random coefficients framework for regional modeling. The framework allows the presence of both the region-specific effects and commonality of responses across regions. Bayes solutions for estimating parameters of interest and for generating predictions are derived. Within a Bayesian framework a predictive density approach to evaluate the impact of changes is suggested. We apply the methodology to evaluate the impact of new rate structures on Ontario regional demand for electricity.