Modeling CO₂ Emissions of Top 5 Emitters with Combination Technique

Abstract:

In this paper, the performance of combination forecast methods for CO₂ emissions prediction is investigated. Linear model, time series model, GM (1, 1) model and Grey Verhulst model are selected in study as the separate models. And, four kinds of combination forecast models, i.e. the equivalent weight (EW) combination method, variance-covariance (VACO) combination method, regression combination (R) method, and discounted mean square forecast error (MSFE) method are chosen to employ for top 5 CO₂ emitters. The forecasting accuracy is compared between these combination models and single models. This research suggests that the combination forecasts are almost certain to outperform the worst individual forecasts and maybe even better than most individual ones. Furthermore the combination forecasts can avoid the risk of model choosing in future projection. For CO₂ emissions forecast with many uncertain factors in the future, combining the single forecast would be safer in such forecasting situations.

• Periodical: Advanced Materials Research (Volumes 616-618)
• Pages: 1143-1147
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.616-618.1143
• Online since: December 2012
• Authors: Wei Sun, Jing Min Wang, Jun Jie Kang
• Keywords: Combination Forecast, Greenhouse Gas (GHG)

References

[1] Working Group III IPCC: IPCC's fourth assessment report (AR4): mitigation of climate change. (Cambridge, United Kingdom 2007).

DOI: 10.1017/cbo9781107415416.002

[2] B. Tutmez: Energ. Explor. Exploit. Vol. 24 (2006), p.139.

[3] Y. Shu and N.S.N. Lam: Atom. Environ. Vol. 45 (2011), p.634.

[4] B. Vincenzo, M. Oronzio and N. Sergio: Energy Vol 34 (2009), p.1413.

[5] S. Pokharel: Energy Policy Vol. 35 (2007), p.350.

[6] Q.M. Liang, Y. Fan and Y.M. Wei: Energy Policy Vol 35 (2007), p.1685.

[7] C.S. Lin, F.M. Liou and C.P. Huang: Appl. Energ. Vol 88 (2011), p.3816.

[8] W.K. Yap and V. Karri: Expert. Syst. Appl. Vol 39 (2012), p.2421.

[9] T. Chen and Y.C. Wang: Intell. Data Anal. Vol 15 (2011), p.763.

[10] J.M. Bates, C.W. Granger: Opl. Res. Q. Vol 20 (1969), p.451.

[11] J.P. Dickinson: Opl. Res. Q. Vol 24 (1973), pp.253-260.

[12] Clive W. J. Granger and R. Ramanathan: Int. J. Forec. Vol 3 (1984), p.197.

[13] C.K. Min and A. Zellner: J. Econometrics Vol 56 (1993) p.89.

[14] M. Hibon and T. Evgeniou: Int. J. Forec. Vol 21 (2005), p.15.

[15] G. P. Zhang: Neurocomputing Vol 50 (2003), p.159.

[16] R.T. Clemen: Int. J. Forec. Vol 5 (1989), p.559.

[17] R.G. Fritz, C. Brandon and J. Xander: Ann. Tourism Res. Vol 11 (1984), p.219.

[18] R. L.Winkler, and S. Makridakis: J. R. Stat. Soc. A Vol 146 (1983), p.150.

[19] BP. 2011. http://www.bp.com/energyoutlook.