Variable weight combination forecasting combines individual forecasting models after giving them proper weights at each time point. Weight is the type of function that changes with forecast time. A relatively rational description of the system can be proposed with the forecasting method, which is of higher precision and better stability. Two individual forecasting models, grey system forecasting and multiple regression forecasting, were generated based on the historical data and influencing factors of coal demand in China from 1981 to 2008. According to the theory of combination forecasting, the variable weight combination forecasting model was formulated to forecast coal demand in China for the next 12 years.
ZHAO Guohao, GUO Shufen, SHENTU Jing, WANG Yongguang
. An Investigation of Coal Demand in China Based on the Variable Weight Combination Forecasting Model[J]. Journal of Resources and Ecology, 2011
, 2(2)
: 126
-131
.
DOI: 10.3969/j.issn.1674-764x.2011.02.004
Chen H Y. 2001. Research on combination forecasting model with non negative and time-variant weights based on effective measure of forecasting methods. Operations Research and Management Science, 1: 50-54.
Dong S Y. 2000. Energy demand Projections based on an uncertain dynamic system modeling approach. Energy Sources, 7: 443-451.
Guo G T. 2004. Analysis and prediction of medium and long term coal supply and demand in China. China Coal, 10: 22-25. (in Chinese)
Jin Q Z, Zhang J. 2004. GM(1, 1Application of the modal in the forecasting of the coal demand. China Coal, 1: 17-19. (in Chinese)
Kouris G. 1998. Energy consumption and economic activity in industrialized countries: a note. Energy Economics, 5: 207-212.
Liu Y S, Wu C. 2008. Forecasting model of coal demand and supply. Journal of Xi’an University of Science and Technology, 3: 7-11. (in Chinese)
Lu E P. 2005. A study of the forecasting modal of Chinese energy demand. Statistics and Decision, 20: 31-33. (in Chinese)
Lu Q, Gu P L, Qiu S M. 2003. The construction and application of combination forecasting model in Chinese energy consumption system. Systems Engineering-theory & Practice, 3: 25-31. (in Chinese)
Renshwa E. 1981. Energy efficiency and the labor productivity in the USA. Energy Eeonomies, 5: 114-152 Tang X W, Zeng Y, Cao C X. 1993. Study of time-variant weight combination forecasts. Forecasting, 3: 48-50. (in Chinese)
Wang J, Liu L D, Wang Z Y. 1997. Current situation and development of combination forecasting model. Forecasting, 6: 7-38. (in Chinese)
Weigen A S. 1996. Time series and prediction using gated experts with application to energy demand forecasts. Applied Artificial Intelligence, 11-12: 583-624.
Xie K G, Zhou J Q. 2000. Research of the weight changeable combination forecast model. Systems Engineering-Theory & Practice, 7: 37-41. (in Chinese)
Xie K G, Zhou J Q. 2003. Combination forecasting models using regression analysis method. Journal of Chongqing University (Natural Science Edition), 1: 64-67. (in Chinese)
Zhao L M, Qian W R, Shen Y F. 2001. Combination of forecasts for demand of dwelling market. Journal of Systems Engineering, 3: 66-69. (in Chinese)
