Title: A combined prediction model of cross-border e-commerce export volume based on BP neural network and SVM

Authors: Haidong Zhong; Jinhui Zhang; Shaozhong Zhang

Addresses: International Economics and Trade Department, School of Economics and Management, NingBo University of Technology, Ningbo, Zhejiang, 315211, China ' E-commerce Department, School of Logistics and E-commerce, Zhejiang Wanli University, Ningbo, Zhejiang, 315100, China ' Engineering of Internet of Things Department, College of Information and Intelligent Engineering, Zhejiang Wanli University, Ningbo, Zhejiang, 315100, China

Abstract: China's cross-border e-commerce (CBEC) is developing rapidly in the last several years and is widely considered as a developing trend of the nation's foreign trade. The prediction of CBEC export volume can effectively reduce the risks, such as goods backlog and long cross-border logistics time, for related enterprises in many ways. However, due to the complex composition of many affecting factors, the prediction accuracy of the most existing methods is usually limited. In the paper, we propose a joint prediction approach that combines the back propagation (BP) neural network model and the support vector machine (SVM) method. A case study with publicly available data in Hangzhou, China, indicates a relative error of less than 1% with the proposed joint predication approach, which is less than the relative error obtained from either BP neural network predication or SVM predication used alone.

Keywords: cross-border e-commerce; export volume predication; BP neural network; SVM algorithm; grey correlation analysis; prediction model; Hangzhou; China; empirical study; causal analysis; joint predication.

DOI: 10.1504/IJTPM.2023.132624

International Journal of Technology, Policy and Management, 2023 Vol.23 No.3, pp.310 - 328

Received: 18 Feb 2022
Accepted: 11 Jun 2022
Published online: 31 Jul 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article