The Comparison About Optimization Methods To Construct Index Investment Portfolio

Index investment is a kind of passive investment strategies. It can be done by investing in all constituents of the index proportional to their share in the index or by selecting a smaller subset of the assets in the index. We can classify the index investment methods into three categories: full replication method, optimized sampling method, and sampling replication method. Seen from the structure principle, stratified sampling method and optimized sampling method are rational and effective. In additional, constructing the optimal index investment portfolio is the most essential link in the two index investment methods. Therefore, the emphasis of this paper is how to construct the optimal index investment portfolio.In this paper, based on the introduction of overseas index optimization methods, we design nine kind of objective functions and some constraint conditions about optimization methods to construct index investment portfolio. Then, in empirical analysis, we choose 180 ingredient stocks of Shanghai Index 180 as the optimized objects, technically construct the optimal index investment portfolio by minimizing nine kind of objective functions to extract the superiorly weight and the superiorly quantity of 180 ingredient stocks, and test the achievements of our portfolio by using out of the sample data. Finally, we choose the two better indexs optimization methods conforming to the actual situation of China's stock market: the minimized-maximized model (MinMax) and the mean-variance optimization method and the whole performance using MinMax is better.