Analysis and forecasts demand for telecom services in Vinh Long in the period 2015-2020

• Main Authors: Loan - Nguyen Thi Anh, 阮氏映灣
• Other Authors: Chun-I Chen
• Format: Others
• Language: en_US
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/72575468254306654999

碩士 === 義守大學 === 管理碩博士班 === 102 === The telecommunication industry is an important service industry of the economy, it creates social product, promotes the development of economic and social. To develop the telecommunication industry in the right direction and planning will create conditions to attract investment, meet the needs of economic and social of Vinh Long province. Beside, it will protect national security, make good economic and social development. Until now, the telecommunication industry quite develop but not synchronous, in Vinh Long. Many places have low quality of service, can not meet the needs of consumers. In current, there is no orientation in planning this industry so the investment in this industry is not inextricably linked to the requirements of economic and social development. To have a good plan, firstable, we must forecast the demand of the people who use telecommunication services in the province. Currently, there are many forecasting methods. However, most methods require data input good and complex, require much time to collect and analyze. Grey model is a forecasting model apply for cases needs the less and simple data input. However, this model also lacks the flexibility to adjust the model to increase the increase the accuracy of forecasts. This thesis will examine the use of Grey Model GM(1,1), Nonlinear Gray Bernoulli Model (NGBM) to forecast the total number of people wishing to use two types of telecommunication services in Vinh Long province. Then forecast results will be used to make a decision on whether to invest in infrastructure to develop that telecommunication services or not. NGBM is a forecasting models which uses less data input, and can be adjusted by selecting optimum n to improve the accuracy of forecasts. This model is particularly good for the kind of non-linear input data.