篇名 | The Prediction of the CO2 Emission of Taiwan by the Grey Prediction Model GDMC(1,n) |
---|---|
卷期 | 2013英 |
作者 | Tzu-Li Tien 、 Shuo-Pei Chen |
頁次 | 162-162 |
關鍵字 | GDMC 、 GMC 、 Convolution integral 、 1-AGO 、 Prediction 、 Greenhouse gas |
出刊日期 | 201312 |
Both grey prediction models GDMC(1,n) and GMC(1,n) are the improved models of the traditional prediction model GM(1,n). Both their grey differential equations are linear differential ones and the accurate modeling values can be obtained because the superposition principle is satisfied by both of them. Only the 1-AGO data of the series characterized by a certain system can be used as the intermediate messages to set up the grey differential equation. The system’s forcing terms are consisted of the 1-AGO data of the associated series by the GMC(1,n) model. In addition to those 1-AGO data, their first order derivatives or even higher order derivatives could affect the system’s behavior. The system’s forcing terms are consisted of the 1-AGO data of the associated series and their first order derivatives by the GDMC(1,n) model. The GDMC(1,n) model is adopted in this paper for raising the accuracy of prediction. The GDMC(1,n) model is used to predict the carbon dioxide emissions of Taiwan in this article. Carbon dioxide is a kind of greenhouse gas. It will absorb the infrared ray and the far infrared ray, and that will raise the atmospheric temperature. Then the temperature of the earth will also increase. The fitting error before the forecast origin will decrease in the model building by adding the additional forcing terms. However, the over-fitting problems will arise in the course of grey model building and the accuracy of prediction by the GDMC(1,n) model. The incompleteness of messages is the primary characteristic of grey theory, and it is necessary to whiten a system by way of inserting more messages in the system. The accuracy of prediction by the GDMC(1,n) model can then be anticipated.