北京邮电大学学报(社科版) ›› 2016, Vol. 18 ›› Issue (6): 31-38.

• 电子商务 • 上一篇    下一篇

基于网络舆情的企业财务危机动态预警

  

  1. 中央财经大学 信息学院,北京10008
  • 收稿日期:2016-08-16 出版日期:2016-12-31
  • 基金资助:
    国家自然科学基金项目(61272398);国家社会科学基金重点项目(13AXW010);北京市哲学社会科学重点项目
    (14JGA001)

Financial Distress Dynamic Warning Based on Online Public Opinion

  1. School of Information, Central University of Finance and Economics, Beijing 100081, China
  • Received:2016-08-16 Online:2016-12-31

摘要:

当前我国企业面临的危机呈现复杂化与多样化的特点,财务信息披露质量问题导致基于财务指标的危机预警模型频频失准,
而引入的非财务指标依赖于特定样本无法适应新危机的特性。为获取全面而客观的企业信息,提出基于自媒体中60家上市公司的相
关评论,获取网民评论纵向数据流,通过情感分析处理形成网络舆情流指标,在此基础上结合财务指标,调节模型时间窗口宽度以
连贯地动态反映企业财务状况,建立基于网络舆情的财务危机动态预警模型。实证表明:引入网络舆情指标的财务预警动态模型可
获取最佳时间窗口宽度,预测效果显著优于仅包括财务指标的静态预警模型,具有更好的适应性。

关键词: 网络舆情, 情感分析, 财务危机预警, 时间窗口

Abstract:

Considering the fact that financial information disclosure quality of listed company cannot be
guaranteed, and non-financial indexes rely on specific samples, a financial warning index system based on online
public opinion is established. In order to obtain full information of enterprises, 60 listed companies in self-
media is collected and sentiment analysis is applied to form the online public opinion index. Then, the financial
dynamic warning system is created by integrating the financial index and online public opinion index with the best
length of time window. The empirical results show that the introduction of financial distress dynamic warning
based on online public opinion leads to the improved accuracy in the financial distress forecast and it is also
more adaptable.

Key words: online public opinion, sentiment analysis, financial distress warning, time window

中图分类号: