Construction of interval-valued forecasting model based on support vector regression and information content of interval

• Main Authors: Wen-Lung Tsai, 蔡文龍
• Other Authors: 呂奇傑
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/73780125962636197605

碩士 === 健行科技大學 === 工業管理系碩士班 === 104 === Forecasting has long been crucial for every company since accurate forecasting results can improve business decision performance. Interval-valued time series forecasting indicates possible future outcomes for upper and lower bounds of interval-valued data and generates a prediction interval. It has the advantage of taking into account the variability and uncertainty so as to reduce the amount of random variation relative to that found in classic point forecasting/ single-valued forecasting model. In this study, a new interval-valued forecasting model based on time-interval information has been proposed. In the proposed forecasting model, a new MSL scheme (mean, standard deviation and level) which is used to describe the bounds of interval-valued data and a support vector regression (SVR) were integrated to construct interval-valued time series forecasting model. A sales data of laptop computer and a Dow Jones stock index data are used as illustrative examples to evaluate the performance of the proposed method. Experimental results showed that the proposed interval-valued time series forecasting model outperforms the naive forecast method, stepwise regression (SR) and integrated SR-SVR method with the new MSL scheme.