基于文本信息的上市中小企业财务困境预测研究

Financial Distress Prediction for Listed SME Based on the Text Information

本文通过网络爬虫获取上市中小企业相关的文本信息,包括以年报为代表的信息披露报告和互联网新闻媒体报道的文本内容,采用词袋方法基于不同特征词词表对这些文本内容进行了量化分析,并以财务变量模型为基础对文本信息量化指标在财务困境预测中的作用进行了实证检验,结果表明由信息披露报告构建的管理层语调变量以及由新闻媒体报道构建的报道倾向变量、负面报道比例变量确实可以提高财务困境模型的拟合度和预测能力,而且在对不同类型文本信息的分析应有不同的侧重点。尽管本文针对的是上市中小企业,但并未考虑市场交易信息,因而可以推广到未上市交易的中小企业。

Massive textual information about the listed small and medium-sized enterprises(SME)has been obtained from the Internet with the web crawler,including information disclosure by the enterprises and online news media reports.All the textual information has been quantified with the bag of words method,and added to the financial distress prediction model based on financial variables.The empirical results show the management tone based on the information disclosure,the media sentiment and the proportion of negative reports based on the online news media reports can be used to improve the fitness and predictive power of the financial distress prediction model.Additionally,different sources of textual information should be analyzed with different lists of words.Although only focusing on listed SME,no market information has been included in the analysis,so that the results can be extended to non-listed SME.