CAJ | 학술논문

【摘要】本文在考虑交易成本和投资组合动态调整的基础上，建立混合整数线性规划模型，引入内核搜索分析框架进行近似求解，并利用沪深300进行实证研究。研究发现，相比于基本内核搜索法，增强型内核搜索法仅在基准指数成分股数量很大时才会较大幅度提高求解质量；考虑投资组合动态调整的模型不仅更稳健，而且跟踪的继承性和保持性更好，尤其适用于单边市场；现实交易特征的过度刻画一定程度上会降低不完全指数复制模型的复制和预测效果。
【적요】본문재고필교역성본화투자조합동태조정적기출상，건립혼합정수선성규화모형，인입내핵수색분석광가진행근사구해，병이용호심300진행실증연구。연구발현，상비우기본내핵수색법，증강형내핵수색법부재기준지수성분고수량흔대시재회교대폭도제고구해질량；고필투자조합동태조정적모형불부경은건，이차근종적계승성화보지성경호，우기괄용우단변시장；현실교역특정적과도각화일정정도상회강저불완전지수복제모형적복제화예측효과。
This paper establishes a mixed-integer linear programming model considering transaction cost and portfolio rebalance, introduces basic and enhanced kernel search for approximate solving and uses the CSI 300 index for empirical re- search. The main conclusions are as follows, firstly, in comparison with basic ker- nel search, only when the number of index constituents is very large, the enhanced method make a substantial improvement in solution quality; secondly, the model considering portfolio rebalance is not only more robust but also better in index- tracking succession, especially is better consideration of real-life features in index cy of replication and prediction. in unilateral market~ thirdly, excessive tracking problem may reduce the efficien-