An Approach To Mid-long Term Energy Demand Forecast And Its Application

Strengthening mid-long term oil demand forecast is vital for controlling air pollution and carbon dioxide emissions.Focusing on modelling mid to long term oil demand,we in this paper summerized historical research on different forecasting techniques and major judgement for China's mid-long term oil demand forecast.Taking IEA' annual forecast as an example,we decompose the leading source of these errors quantitatively.Results suggest that GDP acts as the leading source of demand forecasting errors while fuel price comes thereafter,which requires extra attention in forecasting.Gas,among all fuel types witness the most biased projections.Ignoring the catch-up effect of acquiring rapid economic growth in developing countries such as China will lead to huge mistake in predicting global energy demand.gas appears to be the most severely biased among all fuel types(crude oil as benchmark).Regressions with regional dummy variables consolidate earlier results of China,being outstanding outlier is underestimated more seriously than any other country groups.Opposite trend of overestimation can be seen in Russia demonstrating IEA's relatively higher expectation after the collapse of Soviet Union.We also formulate China's mid-long term oil demand forecasting model detailed to particular kind of oil demand for each industry.Considering different characteristics of productive and non-productve departments,differernt models containing improved hybrid IO model,empirical model,Elastic Coefficient Method and VAR model were used.China's oil demand will reach 650 milion tons in 2020 and peak between 2030 to 2040.Oil demand for 2030 and 2050 will be 730 milion tons and 620 milion tons.Structure of oil demand also vary with oil products for industrial and immerdiate input gaining weight.Transpotaion will be the critical department influencing future oil demand.