基于随机森林方法的投资者概念关注对概念指数收益预测及交易策略的研究  

Research on the Prediction Effect of Investor Attention to the Concept Index on the Return of Concept Index and the Investment Strategy of Concept Index Based on the Random Forest Algorithm

作　　者：扈文秀[1] 苏振兴[1] 杨栎[1] HU Wen-xiu;SU Zhen-xing;Yang Li(School of Economy and Management, Xi’an University of Technology, Xi’an 710054, China)

机构地区：[1]西安理工大学经济与管理学院,陕西西安710054

出　　处：《预测》2021年第1期60-66,共7页Forecasting

基　　金：国家自然科学基金资助项目(71971169,71603203)。

摘　　要：概念炒作一直是股票价格波动的重要驱动因素之一,本文基于投资者关注视角探究投资者概念关注对概念指数波动的预测作用。使用百度搜索指数的非结构化数据测度投资者概念关注,利用随机森林方法构建预测模型,验证市场交易指标和投资者概念关注对概念指数收益和超额收益涨跌的预测作用,并根据预测结果设计交易策略。结果发现市场交易指标对概念指数收益涨跌有较好的预测作用,投资者概念关注则对概念指数超额收益涨跌有较好的预测作用,根据预测结果设计的交易策略可以获得正向的超额收益。本文弥补了前人关于投资者概念关注预测作用研究的空白,也为投资者制定交易策略提供一定的理论参考。The concept speculation has always been one of the main reasons of stock price fluctuation.From the perspective of investor attention,this paper explores the prediction effect of investor attention to the concept index on the fluctuation of concept index.The unstructured data of BSI(Baidu Search Index)is used to measure the investor attention to the concept index,and the Random Forest Algorithm is used to build a prediction model,to verify the prediction effect of market trading indicators and the investor attention to the concept index on the return and abnormal return of concept index.Then,design investment strategies according to the prediction results.The results show that the market trading indicators have a good prediction effect on the return of concept index,and the investor attention to the concept index has a good prediction effect on the abnormal return of concept index,and the investment strategies can get positive abnormal returns.This paper fills in the gap of the existed research on the prediction effect of investor attention to the concept index and provides some academic reference for investors to making investment strategies.

关 键 词：投资者概念关注 概念指数收益 随机森林 交易策略 

分 类 号：F832.5[经济管理—金融学]