Generalized GM (1, 1) model and its application in forecasting of fuel production |
| |
Authors: | Wei Zhou Jian-Min He |
| |
Institution: | 1. International Business School, Yunnan University of Finance and Economics, Kunming 650221, People’s Republic of China;2. School of Economics & Management, Southeast University, Nanjing 211189, People’s Republic of China |
| |
Abstract: | Grey theory is one approach that can be used to construct a model with limited samples to provide better forecasting advantage for short-term problems. Generally, the GM (1, 1) and Discrete GM (1, 1) models are two typical grey forecasting models in grey theory. However, there are two shortcomings in the above grey models respectively, i.e., the homogeneous-exponent simulative deviation in GM (1, 1) model, and the unequal conversion between the original and white equations in DGM (1, 1) model. In this paper, we firstly propose a novel Generalized GM (1, 1) model termed GGM (1, 1) model, based on GM (1, 1) and DGM (1, 1) models, to overcome the above shortcomings. Then, we detailedly study four important properties in this new grey model. Four estimative approaches of stepwise ratio in GGM (1, 1) model context is also covered. In the end, we simulate and forecast the fuel production in China during the period 2003–2010 using three GM (1, 1) models. The empirical results show that GGM (1, 1) model has higher simulative and predictive accuracy than GM (1, 1) and DGM (1, 1) models. This work contributes significantly to improve grey forecasting theory and proposes a optimized GM (1, 1) model. |
| |
Keywords: | GGM (1 1) Stepwise ratio Time responded function Restored function Fuel production |
本文献已被 ScienceDirect 等数据库收录! |
|