GEFCom2014 probabilistic electric load forecasting using time series and semi-parametric regression models

This paper describes the forecasting methodology used by the team ADADA in the load forecasting track of GEFCom2014. The approach includes a semi-parametric regression model for point load forecasting, a multivariate time series simulation model for temperature forecasting, and a resampling strategy for building probabilistic forecasts from point load forecast residuals.