Two-stage Optimization Support Vector Machine for the Construction of Investment Strategy Model

Methods of artificial intelligence have been widely used in the study of investment related topics, and the methods adopted include genetic algorithm and neural network, etc. However, as different to the methods taken in the past, support vector machine is adopted in this article to perform investment strategy study for domestic stock market; investment strategy can be divided into three strategies such as: buy, sell and hold. First, the data was processed, then support vector machine was used to set up investment strategy model, then it was compared with logistic regression for the classification capability of investment strategy. From the empirical results and judging from the classification correctness of four models, it can be seen that the support vector machine after adjustment of input variables and parameters have classification capability relatively superior to that of the other three models.