Abstract
We present an empirical analysis to show that combination of short term load forecasts leads to better accuracy. We also discuss other aspects of combination, i.e., distribution of weights, effect of variation in the historical window and distribution of forecast errors. The distribution of forecast errors is analyzed in order to get a robust forecast. We define a robust forecaster as one which has consistency in forecast accuracy, lesser shocks (outliers) and lower standard deviation in the distribution of forecast errors. We propose a composite ranking (CRank) scheme based on a composite score which considers three performance measures—standard deviation, kurtosis of distribution of forecast errors and accuracy of forecasts. The CRank helps in identification of a robust forecasts given a choice of individual and combined forecaster. The empirical analysis has been done with the real life data sets of two distribution companies in India.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Hahn H, Meyer-Nieberg S and Pickl S 2009 Electric load forecasting methods: Tools for decision making. Eur. J. Oper. Res. 199(3): 902–907
Bates J M and Granger C W J 1969 The combination of forecasts. Oper. Res. Q. 20(4):451–468
Wallis K F 2011 Combining forecasts – forty years later. Appl. Finan. Econom. 21: 33–41
Bichpuriya Y, Rao M S S and Soman S 2010 Combination approaches for short term load forecasting. In: IPEC, 2010 Conference Proceedings, pp 818–823
Bordignon S, Bunn D W, Lisi F and Nan F 2013 Combining day-ahead forecasts for British electricity prices. Energy Econom. 35: 88–103
Shen S, Li G and Song H 2011 Combination forecasts of international tourism demand. Ann. Tourism Res. 38(1): 72–89
Newbold P and Granger C W J 1974 Experience with forecasting univariate time series and the combination of forecasts. J. Royal Stat. Soc. S A (General) 137(2): 131–165
Granger C W J and Ramanathan R 1984 Improved methods of combining forecasts. J. Forecast. 3(2):197–204
Lam K, Mui H and Yuen H 2001 A note on minimizing absolute percentage error in combined forecasts. Comput. Oper. Res. 28(11): 1141–1147
Kiartzis S, Kehagias A, Bakirtzis A and Petridis V 1997 Short term load forecasting using a Bayesian combination method. Int. J. Electr. Power Energy Syst. 19(3): 171–177
Barghinia S, Kamankesh S, Mahdavi N, Vahabie A H and Gorji A 2008 A combination method for short term load forecasting used in Iran electricity market by NeuroFuzzy, Bayesian and finding similar days methods. In: 5th International Conference on European, Electricity Market, 2008. EEM 2008. pp 1–6
Fan S, Chen L and Lee W-J 2009 Short-term load forecasting using comprehensive combination based on multimeteorological information. IEEE Trans. Ind. Appl. 45(4): 1460–1466
DeCarlo L T 1997 On the meaning and use of kurtosis. Psychol. Methods 2(3): 292–307
de Menezes L M, Bunn D W and Taylor J W 2000 Review of guidelines for the use of combined forecasts. Eur. J. Oper. Res. 120(1): 190–204
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Bichpuriya, Y.K., Soman, S.A. & Subramanyam, A. Combining forecasts in short term load forecasting: Empirical analysis and identification of robust forecaster. Sādhanā 41, 1123–1133 (2016). https://doi.org/10.1007/s12046-016-0542-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12046-016-0542-3