Abstract :
Does co-integration help long-term forecasts? We use simulation, real data sets, and multi-step-ahead post-sample forecasts to study this question. Imposing unit-root constraints suggested by co-integration tests produces better forecasts for some cases, but fares poorly for others.The composition of total energy consumption in China is very important, which can be analyzed by means of the co-integration model. Traditional analysis methods of the composition of total energy consumption are compared with co-integration model. The results of empirical research indicate that crude oil, natural gas and hydro-power have the co-integration relationship.It is finally concluded that we can distinguish between a long-run relationship between them. China´s primary energy consumption include coal, crude oil, natural gas and hydro-power, and the crude oil, natural gas and hydro-power have co-integration relationship. In this case, we can distinguish between a long-run relationship between them, we can integrate the series of crude oil, natural gas and hydro-power. This would include, for example, fuel oil, supply of electricity, and raw materials.
Keywords :
coal; energy consumption; hydraulic systems; hydroelectric power; natural gas technology; power engineering computing; program testing; total energy systems; coal; cointegration model; crude oil; hydro-power; long-term forecast; multistep-ahead post-sample forecast; natural gas; software testing; total energy consumption; unit-root constraint; Computer science; Content addressable storage; Energy consumption; Energy management; Load forecasting; Natural gas; Petroleum; Predictive models; Software engineering; Software testing; co-integration; energy consumption; long-term forecast; time series;
