文章摘要
孟勇,任梦,赵心.行业资产的Black-Litterman模型配置研究[J].数量经济技术经济研究,2022,(1):154-173
行业资产的Black-Litterman模型配置研究
Research on Black-Litterman Model Allocation of Industry Asset
  
DOI：
中文关键词: 投资者情绪  Black-Litterman模型  行业资产配置  文本挖掘
英文关键词: Investor Sentiment  Black-Litterman Model  Industry Asset Allocation  Text Mining
基金项目:本文获得国家社会科学基金项目“超高维稀疏网络模型及其组合风险应用研究”(19BTJ026)资助。
作者单位
孟勇 山西财经大学统计学院 
任梦 山西财经大学财政与公共经济学院 
赵心 山西财经大学金融学院 
中文摘要:
      研究目标：构建反映行业股价走势的基于社交网络文本挖掘算法的行业投资者情绪指标，并改善嵌入行业投资者情绪指标的Black-Litterman模型对资产的配置结果。研究方法：基于社交网络文本挖掘算法度量投资者情绪，运用主成分分析法构建行业投资者情绪指标，并嵌入Black-Litterman模型中构建投资者观点矩阵，确定行业资产配置比。研究发现：基于行业投资者情绪的BL模型有效提高了资产配置的日均收益率和夏普比率。实证结果在样本外验证（除受新冠疫情影响阶段）、暴涨暴跌阶段以及经过允许卖空和交易成本调整后仍稳健，进而证实了投资者情绪对资产组合有显著影响。研究创新：基于社交网络文本挖掘算法构建投资者情绪指数，解决了仅依赖于预期收益或历史数据的预测模型无法直观揭示投资者心理认知和行为的局限性问题，从一个崭新的视角科学地解决Black-Litterman模型中投资者观点的生成问题。研究价值：扩展了Black-Litterman模型理论体系研究，并推动了行为金融理论在资产配置中的应用。
英文摘要:
      Research Objectives: Construct industry investor sentiment indicators based on social network text mining technology that reflects the trend of industry stock prices, and improve the asset allocation results of the Black-Litterman model embedded in industry investor sentiment indicators. Research Methods: Based on social network text mining technology and principal component analysis, this paper constructed investor sentiment indicators of the industry which forms the investor views matrix of the BL model, and determined the asset allocation ratio of the industry. Research Findings: The BL model based on industry investor sentiment can effectively improve the average return rate and Sharpe ratio of asset allocation. Such findings are robust to the out-of-sample verification (except for the stage of COVID-19), the stage of sharp rise and fall, short selling and transaction cost adjustments. Research Innovations: The construction of investor sentiment index based on social network text mining technology solves the limitation that predictive models that only rely on expected returns or historical data cannot visually reveal the psychological cognition and behavior of investors. Scientifically solve the problem of generating investors opinion in the Black-Litterman model from a new perspective. Research Value: This paper expands the research of Black-Litterman model theory system and promotes the application of behavioral finance theory in asset allocation.
查看全文