登录 EN

添加临时用户

流动性风险与加密数字货币定价

Liquidity Risk and Cryptocurrency Pricing

作者:崇辰钊
  • 学号
    2022******
  • 学位
    硕士
  • 电子邮箱
    Joe******com
  • 答辩日期
    2024.05.10
  • 导师
    何平
  • 学科名
    金融
  • 页码
    64
  • 保密级别
    公开
  • 培养单位
    051 经管学院
  • 中文关键词
    加密数字货币;流动性风险;资产定价;多因子模型
  • 英文关键词
    Cryptocurrency;Liquidity Risk;Asset Pricing;Multi-factor Model

摘要

资产定价是金融学中一个经久不衰的核心课题,对股票、债券等成熟资产的定价理论和实证分析数不胜数,贡献了金融学领域中一系列标志性成果。加密数字货币作为近年来新兴的一类可投资资产,在资产属性、市场有效性、投资者行为等维度和股票、债券等资产具有显著区别,关于加密数字货币的资产定价理论和实证研究尚处于快速发展阶段。目前关于加密数字货币的定价研究中,Liu et al.(2022)提出的由市场因子、市值因子、动量因子组成的加密数字货币三因子定价模型对加密数字货币的收益率有较好的解释效果。流动性风险是投资加密数字货币时所面临的一类重要风险,为了探究流动性风险对加密数字货币市场资产定价的影响,本文选取了2019年1月至2024年2月的加密数字货币市场交易数据,构建了由市场因子、市值因子、动量因子、流动性因子组成的加密数字货币市场四因子定价模型,并与单因子模型、三因子模型进行对比,研究不同模型对加密数字货币收益率的解释能力。通过对因子的描述性统计、相关性分析和Fama-Macbeth回归检验发现,流动性因子具有显著的超额收益。使用市场因子、市值因子、动量因子作为解释变量,对流动性因子收益率进行回归分析,发现流动性因子具有上述三因子之外的显著超额收益。这些分析结果表明,加密数字货币市场存在较为显著的流动性溢价现象,流动性差的加密数字货币资产提供了更高的超额收益。通过比较不同模型的GRS统计量、平均调整R方、截距项绝对值的平均值以及截距项显著个数,本文发现,在上述各评价维度上,加入流动性因子后的加密数字货币四因子定价模型对加密数字货币收益率的解释能力强于单因子模型和三因子模型。

Asset pricing is an enduring core topic in finance. There are countless pricing theories and empirical analyzes of mature assets such as stocks and bonds, and they have contributed a series of landmark results. Cryptocurrency, as an emerging class of assets, is significantly different from stocks and bonds. The asset pricing theory and empirical research on cryptocurrency is still in development. In current research, the three-factor pricing model of cryptocurrency proposed by Liu et al. (2022), consisting of market factor, market capitalization factor, and momentum factor, has a good explanation effect on the return of cryptocurrency.Liquidity risk is an important risk faced by cryptocurrency investors. In order to explore the impact of liquidity risk on asset pricing in the cryptocurrency market, this article selected the cryptocurrency market transaction data from January 2019 to February 2024 to construct a four-factor pricing model consisting of market factor, market capitalization factor, momentum factor, and liquidity factor, and compared the performance in explaining cryptocurrency returns with the single-factor model and the three-factor model.Through the descriptive statistics, correlation analysis and Fama-Macbeth regression test of the factors, it is found that the liquidity factor has significant excess returns. Using the market factor, market capitalization factor, and momentum factor as independent variables, a regression analysis was conducted on the liquidity factor, and it is found that the liquidity factor had significant excess returns in addition to the above three factors. These analyses show that cryptocurrency market has a significant liquidity premium phenomenon, and illiquid cryptocurrency assets provide higher excess returns.By comparing the GRS statistic, the average adjusted R-squared, the average of the absolute values of the intercept terms, and the number of significant intercept terms, this article found that in all above evaluation dimensions, the four-factor pricing model‘s performance is better than the three-factor model and the single-factor model.