Analysis of Regional Logistics Demand Prediction Based on Support Vector Non-Linear Multiple Regression

This paper indicates the implications and characteristics of the regional logistics demand, and the key prediction indicator system in the course of regional logistics demand prediction is established and proposes an approach that support vector multiple regression (SVR) and principal component analysis (PCA) can solve the prediction problem of regional logistics demand. A numerical example based on Heilongjiang province of China is given to demonstrate the effectiveness and practicality of the proposed approach.