摘要
将技术指标引入决策树的构建,并结合Adaboost模型优化,以上证综指为例提出一种交叉验证的股票指数价格涨跌预测模型.基于预测模型构建对时间序列数据的量化择时系统,并评估其策略表现.实证结果表明,本文提出的预测模型精度较高,基于预测模型的量化择时系统有更好的表现,具有较优秀的市场状况识别状态和获得超额收益的能力.该结果为决策树在择时策略中的创新性运用提供参考,为技术指标的创新运用提供新的思路.
In this paper,technical indicators are introduced into the construction of the decision tree and combined with the optimization of the Adaboost model,a cross-validated stock index price uctuation prediction model is proposed for the Shanghai Composite Index as an example.Based on the prediction model,a quantitative timing system for time series data is constructed,and its strategic performance is evaluated.The empirical result shows that the prediction model proposed in this paper has higher accuracy,and the quantitative timing system based on the prediction model has better performance and ability to identify market conditions and obtain excess returns.The result provides a reference for the innovative application of decision trees in timing strategy and provides new ideas for innovative applications of technical indicators.
作者
刘向丽
崔文泓
LIU Xiangli;CUI Wenhong(Department of Finance,Central University of Finance and Economics,Beijing 102206,China;IEOR Department,University of California at Berkeley,Berkeley 94720,United States)
出处
《纯粹数学与应用数学》
2023年第3期339-349,共11页
Pure and Applied Mathematics
基金
国家自然科学基金(71471182).
关键词
决策树
量化择时
交易策略
股票指数
decision tree
quantitative timing
trading strategies
stock indices
