Online Program Home
  My Program

Abstract Details

Activity Number: 165 - Statistics for Business and Financial Markets
Type: Contributed
Date/Time: Monday, July 31, 2017 : 10:30 AM to 12:20 PM
Sponsor: Business and Economic Statistics Section
Abstract #324385
Title: Automatic Forecasting of Hourly Electricity Demand with a Computationally Efficient Semiparametric Time Series Model
Author(s): Jun Liu*
Companies: Georgia Southern University
Keywords: Electricity ; Forecasting ; Nonparametric Regression ; Time Series ; ARIMA

In this paper we develop a semi-parametric approach to model nonlinear relationship in time series data. The usefulness of this approach is illustrated on a hourly electricity demand data set. Polynomial splines are used to model the effect of temperature on hourly electricity demand for different times of the day and types of the day. An ARIMA model is used to model the serial correlation in the data. An algorithm is developed to automatically select the models, and the models are estimated through backfitting. Forecasting performance is evaluated using post-sample forecasting and comparative results are presented.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

Copyright © American Statistical Association