Lee, Jun-yaw creator Sauter-Servaes, Florian Chairperson I, Fude Examination Committee Lucas, N.J.D. Examintion Committee Government of Republic China Scholarship Donor text th Bangkok Asian Institute of Technology 1984 monographic eng 110 p. This study presents three different approaches to electricity demand forecasting and applied them to the case of Taiwan up to the year 1992. The econometric models are estimated for each of eight economic sectors separately, using yearly data from 1970 to 1981 for each sectors. The price of electricity, the price of natural gas and value added in each sector are important factors determining the electricity demand. Long and short term elasticities are estimated and discussed. The model forecasts electricity demand by different scenarios. A probability forecast is also made to include the stochastic nature of the different scenarios in the considerations. The Box-Jenkins ARIMA and bivariate transfer function models are also used to compare the performance of these methods. It is concluded that the bivariate transfer function model could be a very good approach to the aggregate electricity demand forecast. The forecasting results show that electricity demand will continue to grow. However, the rates of growth will be considerably lower than those observed in the 1960s and 1970s. A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Environment, Resources & Development Thesis (M.Eng.) - Asian Institute of Technology, 1984 Electricity Taiwan Asian Institute of Technology. http://203.159.5.9/ait-thesis/detail.php?q=B20844 http://203.159.5.9/ait-thesis/detail.php?q=B20844 250998 20260818095502.0