首页|机构投资者交易与债券收益率——基于银行间债券市场的理论与实证研究

机构投资者交易与债券收益率——基于银行间债券市场的理论与实证研究

扫码查看
本文基于中国银行间债券市场的日度交易数据,构建以订单流为基础的交易指标,通过包含公开信息和私有信息的债券定价模型以及时间序列聚类的投资组合分析和多种模型分析,对各类机构投资者预测债券收益率变化的能力进行研究.研究发现,在银行间债券市场,保险机构、公募基金、私募基金和境外机构的交易能够预测债券收益率的变化,但在交易券种(国债、政策性银行债)、预测期限、预测能力来源等方面均存在显著差异.本文还对机构投资者的预测能力进行了机制检验.研究发现,在境内机构投资者中,公募基金具备更强的交易预测能力和信息获取与解读能力;境外机构投资者对宏观货币政策信息具有更强的预测能力.本文的研究结果表明,在我国银行间债券市场,机构投资者的交易有助于促进价格发现,增加债券收益率的信息含量,从而提升市场有效性.最后,本文基于实证结果提出相关政策建议.
Institutional Trading and Bond Yields——Theoretical and Empirical Research from the Interbank Bond Market
Based on the daily transaction data of China's interbank bond market,this paper constructs a transaction indicator based on order flows,and studies the ability of different types of institutional investors to predict changes in bond yields through bond pricing mod-els incorporating both public and private information,time series clustering portfolio analy-sis and multiple model analysis.The paper finds that:in the interbank bond market,the transactions of insurance companies,public equity funds,private equity funds,and for-eign institutional investors can predict changes in bond yields,but institutional investors'predictive ability differs in bond types(government bonds and financial bonds),prediction horizons,and sources of the predictive ability.This paper also studies the mechanism be-hind the predictive ability of institutional investors and finds that:among domestic institu-tional investors,public equity funds are better at predicting trading,as well as acquiring and interpreting information;foreign institutional investors have a stronger ability to pre-dict macro and monetary policy information.These results show that in China's interbank bond market,institutional trading contributes to price discovery,increases the information content of bond yields,and improves market efficiency.Finally,this paper puts forward relevant policy recommendations based on the empirical results.

BondsInstitutional investorsTrading behaviorReturn predictability

唐彬、蒋宇翔、肖钰、杨粲

展开 >

西南财经大学中国金融研究院(邮编:611130)

中央国债登记结算有限责任公司(邮编:100033)

中央财经大学(邮编:100081)

债券 机构投资者 交易行为 收益率可预测性

2024

南开经济研究
南开大学经济学院

南开经济研究

CSTPCDCSSCICHSSCD北大核心
影响因子:1.82
ISSN:1001-4691
年,卷(期):2024.(4)
  • 36