• 统计研究中心
当前位置: 首页> 系列讲座> 正文

香港大学朱柯副教授:Empirical asset pricing via the conditional quantile variational autoencoder

主 题:Empirical asset pricing via the conditional quantile variational autoencoder

主讲人:香港大学朱柯副教授

主持人:统计学院林华珍教授

时间:2022年7月27日(周三)上午10:30-11:30

直播平台及会议ID:腾讯会议,ID: 105-703-908

主办单位:统计研究中心和统计学院 科研处


主讲人简介:

朱柯博士2011年获得香港科技大学统计学博士学位,同年进入中国科学院数学与系统科学研究院从事研究工作,历任助理研究员、副研究员。2016年加入香港大学任助理教授、副教授。朱柯博士的研究兴趣包括统计建模、金融时间序列分析、计量经济、金融大数据、因果推断等领域。朱柯博士2015年获得中国科学院数学与系统科学研究院的 “陈景润未来之星奖”,他的一系列论文发表在国际统计学和计量经济学顶级杂志 Annals of Statistics, Journal of the Royal Statistical Society (Series B), Journal of American Statistics Association, Journal of Econometrics 和 Journal of Business & Economic Statistics。


内容提要:

We propose a new asset pricing model that is applicable to the big panel of return data. Our model aims to explain the conditional mean of the return from the conditional distribution of the return, which is approximated by a step distribution function constructed from conditional quantiles of the return. To study conditional quantiles of the return, we propose a new conditional quantile variational autoencoder (CQVAE) network. The CQVAE network specifies a factor structure for conditional quantiles with latent factors learned from a VAE network and nonlinear factor loadings learned from a "multi-head" network. Under the CQVAE network, we allow the observed covariates such as asset characteristics to guide the structure of latent factors and factor loadings. Furthermore, we provide a two-step estimation procedure for the CQVAE network. Finally, we apply our CQVAE asset pricing model to analyze a 60-year US equity return data set. Compared with the benchmark conditional autoencoder model, the CQVAE model not only delivers much larger values of out-of-sample total and predictive R^2s, but also earns at least 30.9% higher values of Sharpe ratios for both long-short and long-only portfolios.


上一条:中国科学院张新雨研究员:Optimal parameter-transfer learning by semiparametric model averaging

下一条:东北师范大学郑术蓉教授:Central limit theorems of high-dimensional sample correlation matrix under two stuctures