Skip to Main Content
September 16, 2014
Brattle Economists to Present Webinar for the National Association of Regulatory Utility Commissioners on a New Method of Forecasting Peak Demand

Brattle principal Ahmad Faruqui and associate Charlie Gibbons will be presenting the webinar, "New Method of Forecasting Peak Demand,” to the National Association of Regulatory Utility Commissioners (NARUC) on September 16, 2014.

During the webinar, Drs. Faruqui and Gibbons will discuss the new method of forecasting peak demand, quantile regression, which has been shown to be more accurate than the more commonly used method, ordinary least squares (OLS). This is according to a recent Brattle study, which demonstrates how OLS approaches mostly characterize peak demand days by extreme values of predictors (such as weather), causing OLS to produce biased estimates of the annual peak. Brattle’s new estimation procedure, optimal forecast quantile regression (OFQR), uses quantile regression to predict the annual peak. In this study, Brattle compares the results of the OLS and OFQR estimation approaches for 32 utility zones, finding the OFQR approach unbiased and the OLS under-forecasts by nearly 5% on average.

For more information on OFQR, please visit the Social Science Research Network website to download this report by Drs. Faruqui and Gibbons.