Grey Support Vector Autoregression Model with Application to iPhone Demand Forecasting

摘要

The maturity of 3G system contributes in making Smartphones as a mainstream product in the mobile phone market. Among the Smartphone products, iPhone is the most attractive mobile phone. As a result, an accurate demand forecasting of iPhone is important for its competitors in the market. Because of continuous innovation and increasing customers' aspiration for various new functions and apps, the demand for iPhone is dramatically increasing. Consequently, it is difficult to forecast iPhone's demand by using the historical data. In order to make a precise forecasting of the demand for iPhone, we used the support vector regression model in function approximation based on the structural risk minimization, the autoregressive model in which the output variable depends linearly on its own previous values, and grey theory with limited data to construct a new grey support vector autoregression (GSVAR) model. With these we could obtain a better forecasting result. Finally, we illustrated an application of the best forecasting process in predicting the demand for iPhone in the market.