Medium-term electricity demand forecasting based on MARS
Authors:
- 1 September 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Abstract
The paper proposes use of multivariate adaptive regression splines (MARS) method to perform monthly electricity demand forecasting for medium-term. The model is developed based on specific example of Turkey; however is applicable to any other system. Performance of the proposed method is compared to that of multiple linear regression (MLR), generalized additive model (GAM), and artificial neural networks (ANN) methods. The validation process shows that the proposed model outperforms the other ones by test error and shows stable error performance.Keywords
This publication has 16 references indexed in Scilit:
- Mid-long term Algerian electric load forecasting using regression approach2013
- A comparison of support vector machines and artificial neural networks for mid-term load forecasting2012
- Medium-term load forecasting using neural network approach2011
- A variable structure neural network model for mid-term load forecasting of Iran national power system2008
- Comparative performance of generalized additive models and multivariate adaptive regression splines for statistical modelling of species distributionsEcological Modelling, 2006
- A Knowledge Based System for Medium Term Load Forecasting2006
- Forecasting the hourly Ontario energy price by multivariate adaptive regression splines2006
- Analyzing the Impact of Weather Variables on Monthly Electricity DemandIEEE Transactions on Power Systems, 2005
- Market Operations in Electric Power Systems2002
- Multivariate Adaptive Regression SplinesThe Annals of Statistics, 1991