本文研究了指数型基金管理和指数套利中最核心的指数追踪问题。依据结构风险最小化思想,建立了基于支持向量机的指数追踪模型,并利用OR-Library中的测试数据之恒生指数历史数据进行了实证检验。数据结果表明本文提出的新方法能够提高样本外的追踪效果,同时也说明该方法具有良好的鲁棒性,从而具有较高的理论和实用价值。
In this paper, we study the index tracking problem which plays a core role in index fund management and index arbitrage trade. To minimize structural risks, we propose an SVM Based index tracking model, and test it with the historical data of Hang Seng index from OR-Library. The numerical results show that our method can improve the out-ofsample performance and ensure a robust tracking portfolio, thus demonstrating both theoretical and practical value.