基于双重选择LASSO模型的我国股市定价因子边际有效性研究  

Research on the marginal effectiveness of Chinese stock markets’pricing factors: Application of double-selection LASSO model

作　　者：毛杰 陈宓舟[3,4,5] 许磊 杜树楷 MAO Jie;CHEN Mizhou;XU Lei;DU Shukai(School of Economics,Shanghai University,Shanghai 200444,China;Financial Research Center,Fudan University,Shanghai 200433,China;Postdoctoral Workstation,Orient Securities Co.,Ltd.,Shanghai 200010,China;Financial Products Division,Orient Securities Co.,Ltd.,Shanghai 200010,China;School of Finance,Shanghai University of Finance and Economics,Shanghai 200433,China)

机构地区：[1]上海大学经济学院,上海200444 [2]复旦大学金融研究中心,上海200433 [3]东方证券股份有限公司博士后工作站,上海200010 [4]东方证券股份有限公司金融产品总部,上海200010 [5]上海财经大学金融学院,上海200433

出　　处：《系统工程理论与实践》2024年第9期2993-3005,共13页Systems Engineering-Theory & Practice

基　　金：国家社会科学基金重大项目(20&ZD102);上海高水平地方高校创新团队项目。

摘　　要：在高维数据背景下,传统的因子定价估计方法可能无法准确判断定价因子的有效性.鉴此,本文构建了双重选择LASSO模型,估计了定价因子的随机贴现因子载荷,以替代估计风险溢价的传统因子估计方法,籍以在高维数据背景下准确判断出定价因子的边际有效性.本文继而收集了85个资产定价因子,构建了我国股市的高维定价因子库,并发现在2014年之后发现的15个定价因子中有7个定价因子是边际有效定价因子.此研究结论在多种稳健性检验下都基本保持一致,由此多番验证了本文实证方法的稳健性.通过进一步分析,本文还在时变随机贴现因子的情况下发现了上述定价因子的有效性基本保持一致.In the present era of high-dimensional data,it is quite unlikely that the traditional methods for estimating pricing factors are capable of judging accurately the marginal effectiveness of pricing factors applicable to the Chinese stock markets.Hence we construct a double-selection LASSO model,instead of the traditional methods which estimate stock pricing factors mainly by estimating risk premium.And then by means of this double-selection LASSO model,we estimate stochastic discount factors loading,thereby being able to judge accurately the marginal effectiveness of stock pricing factors while processing high-dimensional data.Next,we gather together 85 pricing factors applicable to the Chinese stock markets,thus building up a highdimensional pricing factor zoo.In addition,we identify 7 marginally effective pricing factors out of the 15 factors discovered after 2014.Our discovery proves consistent in various robustness tests. We further find in the follow-up analysis that the effectiveness of these 7 pricing factorsproves consistent under the conditions of time-varying SDF.

关 键 词：定价因子边际有效性 双重选择LASSO模型 随机贴现因子 机器学习 因子库 

分 类 号：F830.91[经济管理—金融学] F830.59