paper:empirical_asset_pricing_via_machine_learning
差别
这里会显示出您选择的修订版和当前版本之间的差别。
两侧同时换到之前的修订记录前一修订版后一修订版 | 前一修订版 | ||
paper:empirical_asset_pricing_via_machine_learning [2019/09/09 15:47] – kk | paper:empirical_asset_pricing_via_machine_learning [2023/11/10 12:13] (当前版本) – 外部编辑 127.0.0.1 | ||
---|---|---|---|
行 5: | 行 5: | ||
http:// | http:// | ||
+ | |||
+ | ====== Empirical Asset Pricing via Machine Learning====== | ||
+ | |||
+ | 对本文的总体评价为:4 | ||
+ | |||
+ | 可参考以下标准: | ||
+ | |||
+ | * 5分:佳作、开创性成果 | ||
+ | * 4分:合格的优秀论文、可直接接收发表 | ||
+ | * 3分:小改(Minor)后可接收 | ||
+ | * 2分:需要大改(Major) | ||
+ | * 1分:价值有限,即使修改后亦不能发表 | ||
+ | * 0分:本wiki不收录0分的论文 | ||
+ | |||
+ | ===== 文献基本信息 ===== | ||
+ | |||
+ | ==== 标题 ==== | ||
+ | Empirical Asset Pricing via Machine Learning | ||
+ | ==== 作者 ==== | ||
+ | - Shihao Gu,Booth School of Business-University of Chicago | ||
+ | - Bryan Kelly,Yale University, AQR Capital Management, and NBER | ||
+ | - Dacheng Xiu,Booth School of Business University of Chicagohina | ||
+ | |||
+ | |||
+ | ==== 出版年份 ==== | ||
+ | 2019 | ||
+ | |||
+ | ==== 来源 ==== | ||
+ | SSRN | ||
+ | |||
+ | ==== 关键词 ==== | ||
+ | Machine Learning, Big Data, Return Prediction, Cross-Section of Returns, Ridge Regression, (Group) Lasso, Elastic Net, Random Forest, Gradient Boosting, (Deep) Neural Networks, | ||
+ | Fintech | ||
+ | ==== 摘要 ==== | ||
+ | We synthesize the field of machine learning with the canonical problem of empirical asset pricing: measuring asset risk premia. In the familiar empirical setting of cross section and time series stock return prediction, we perform a comparative analysis of methods in the machine learning repertoire, including generalized linear models, dimension reduction, boosted regression trees, random forests, and neural networks. At the broadest level, we find that machine learning offers an improved description of expected return behavior relative to traditional forecasting methods. We identify the best performing methods (trees and neural networks) and trace their predictive gains to allowance of nonlinear predictor interactions that are missed by other methods. Lastly, we find that all methods agree on the same small set of dominant predictive signals that includes variations on momentum, liquidity, and volatility. Improved risk premium measurement through machine learning can simplify the investigation into economic mechanisms of asset pricing and justifies its growing role in innovative financial technologies. | ||
+ | ==== 引用方式 ==== | ||
+ | 建议使用Google Scholar中的Chicago版式 | ||
+ | |||
+ | Diallo, Boubacar, Aliyu Bagudu, and Qi Zhang. "A Machine Learning Approach to the Fama-French Three-and Five-Factor Models." | ||
+ | |||
+ | ==== 链接 ==== | ||
+ | https:// | ||
+ | |||
+ | |||
+ | ===== 评阅意见 ===== | ||
+ | |||
+ | ==== 文献简介 ==== | ||
+ | |||
+ | 本篇论文将机器学习和资产定价的典型问题——测量资产的风险溢价——进行了综合。文章选用了多种具有代表性的机器学习方法(包括线性回归、带惩罚的广义线性模型, | ||
+ | |||
+ | |||
+ | |||
+ | ==== 文献评价 ==== | ||
+ | |||
+ | === 创新性 === | ||
+ | 本篇文章最大的创新点是将各个机器学习方法做了横向比较(包括与传统预测方法之间的比较),并将机器学习应用在两个典型资产定价问题上(横断面预测收益和时间序列预测收益)并进行了比较。 | ||
+ | |||
+ | === 相关性 === | ||
+ | 文章通过比较发现了最具有预测价值的股票特征指标,为风险溢价方面的预测提供了一套基准,并且通过横向比较各个机器学习方法对组合预测的能力,找出了这些方法中预测能力最好的机器学习方法。 | ||
+ | |||
+ | |||
+ | === 严谨性 === | ||
+ | 首先在比较分析最有价值的股票特征指标时,通过不同的方法分析,得出相同的结果;其次在股票特征指标的样本中加入了5个“安慰剂”指标,同样得出了相同的结果,证明了结论的稳健性。在寻找最优机器学习方法时,通过构建三种不同的投资组合,同样得到相同的结果,再次证明了结论的稳健性。 | ||
+ | ==== 需改改进之处 ==== | ||
+ | |||
+ | |||
+ | |||
+ | |||
+ | ==== 需要小改的地方 ==== | ||
+ | |||
+ | |||
+ | |||
+ | ==== 进一步研究的可能与方向 ==== | ||
+ | |||
+ | 1、 可以尝试加入强化学习的方法进行验证,检验强化学习方法的优劣; | ||
+ | 2、 可以使用文中选出的最有价值的股票特征指标,进行资产定价; | ||
+ | |||
+ | ==== 其他评价 ==== | ||
+ |
paper/empirical_asset_pricing_via_machine_learning.1568015234.txt.gz · 最后更改: 2023/11/10 12:12 (外部编辑)