This paper studies prediction to the energy consumption structure composed by the proportions of the coal, petroleum, natural gas and new energy resources of China by building compositional time series model based on the historical data from 1978 to 2008. Compared to the traditional statistical methods, compositional data analysis can effectively solve the practical difficulties encountered by the unit sum constrained data during its modeling process, and can better disclose the integral and associated characteristics of the proportions. Empirical study shows that in the short-term future of China, the coal resource will remain the main consumption energy with the highest proportion in the energy consumption structure; the proportions of the petroleum and natural gas are far lower than the world level and have remarkable differences with those of the major countries in the world; the proportion of the new energy obviously keeps rising and need more exploiting. It is hoped that the research work can further enrich and extend the achievements in the energy structure application field.
