基于投资者情绪的创业板股票收益率预测研究

Research on GEM Stock Return Forecasting Based on Investor Sentiment

以创业板股票市场为主要研究对象,基于文本挖掘方法对创业板股票收益率进行预测分析,利用词典法对从东方财富股吧爬取的2021年4月1日至2023年4月1日创业板股票评论的情感倾向进行分类,建立投资者情绪指数,构建基于粒子群算法优化的支持向量机(particle swarm optimization support vector machine,PSO-SVM)模型对收益率进行预测分析。在实证分析阶段,以创业板中流通市值最大的股票——“宁德时代”为代表,利用PSO-SVM模型对其收益率进行预测分析,同时设置一系列对照模型进行对比分析。结果表明:提出的模型预测结果优于其他对照组模型(多元线性回归、随机森林、支持向量机),而引入情绪指数的模型预测效果比未引入情绪指数的模型预测效果更好。

Taking the GEM stock market as the main research object and the prediction of GEM stock returns was analyzed based on text mining method.The lexicon method was used to classify the sentiment tendency of GEM stock comments crawled from the Oriental Fortune stock bar from April 1,2021 to April 1,2023 to build an investor sentiment index.A support vector machine(Particle Swarm Optimization Support Vector Machine,PSO-SVM) model optimized based on particle swarm algorithm was constructed to predict the return.In the empirical analysis stage,“Ningde Times”,the stock with the largest outstanding market capitalization in GEM,was selected as a representative,and its return was predicted and analyzed by PSO-SVM model,and a series of control models were set up for comparative analysis at the same time.The results show that the model proposed in this paper is better than the other control models(multiple linear regression,random forest,support vector machine),and the model with the introduction of the sentiment index is better than the model without the introduction of the sentiment index.