Research on Forecasting Model for Logistics Demand Based on Support Vector Machine

摘要

The paper analyzes the importance of logistics demand of forecasting and introduces the main forecasting methods both at home and abroad. Then the paper sums up core ideas and basic theories of Support Vector Machine and builds a forecasting model of logistics demand using this new theory. The paper also expounds on the analytic and applicable process of the model with the process of the parameters of calibration and correction. Thirdly, this paper builds on Beijing's logistics demand forecasting model based on the Support Vector Machine and applies LibSVM software to calculate the results. The results test the accuracy and feasibility of the model and indicate that it has a better utility value than previously supposed.