Modelling Residential and Commercial Demand for Electricity Using Autoregressive Distributed Lag Models

Kenneth H. Tiedemann


Time-series modelling, Regression analysis, Identification, Energy efficiency


Utilities, utility regulators and governments need detailed and timely information on energy demand to support planning and development of generation, transmission and distribution investments in a prudent, timely and cost effective manner, and information on price and income elasticities of demand is particularly important. Since electricity demand elasticity estimates vary significantly across jurisdictions, it is important to have jurisdiction-specific elasticity estimates. This study helps fill the gap for British Columbia by estimating sector electricity demand functions using autoregressive distributed lag econometric models. The study has three main conclusions. First, the estimated partial adjustment demand models have good statistical properties in the sense that explanatory power is high, the regression coefficients are statistically significant, and the level of autocorrelation is minimal. Second, electricity demand in British Columbia for the residential and commercial customer segments exhibits limited responsiveness to changes in electricity prices or natural gas prices but high responsiveness to the level of economic activity. Third, the own-price elasticity estimates for British Columbia are consistent with the most recent detailed regional level information available for the United States.

Important Links:

Go Back