Welcome to my personal website.

— Yaohan Chen —(陈垚翰)

— Yaohan Chen —(陈垚翰)

Assistant Professor
School of Big Data and Statisitics
Anhui University


I conduct research on Bayesian econometrics and machine learning methods in empirical asset pricing. In particular, I am interested in developing and applying Bayesian methods or machine learning methods from the Bayesian perspective for solving the problems in empirical asset pricing and macroeconomics.

Before obtaining my PhD from Singapore Management University, I graduated with a Bachelor’s degree in mathematics from Central China Normal University, and a Master’s degree in economics from Wuhan University in China. I am enthusiastic about computer programming and developing software packages for solving academic problems in economics and finance. Aside from academia, I enjoy reading books about history and Chinese literature. I also like playing table tennis occasionally in my leisure time.


  • Bayesian Econometrics
  • Machine Learning (Application from Bayesian Perspective)
  • Empirical Finance (Asset Pricing)


  • PhD in Economics, 2022

    Singapore Management University

  • Master in Economics, 2017

    Wuhan University

  • BSc in Mathematics, 2014

    Central China Normal University


Working Paper
  • Alternative Parametric Models for Spot Volatility in High Frequency: A Bayesian Approach (2022) (Job Market Paper)
    [PDF] [Code]
    • Novel correction for recently proposed methodology for estimating spot volatility.
    • Efficient C++ code for implementation is currently available upon request.
  • Sparse Structure of Stochastic Discount Factor in the Chinese Stock Market: A Bayesian Interpretable Machine-learning Approach (2021)
    [PDF] [Data] [Code]
    • I review an interpretable machine learning method more in technical details from the Bayesian perspective.
    • Empirically, I extend the application of this method to investigate the SDF structure in the Chinese stock market.
    • Data for the firm-level characteristics in the Chinese stock market is available upon request.
  • Do Volatility-Managed Portfolios Work? Empirical Evidence from the Chinese Stock Market (2021)
    [PDF] [Data] [Code]
    • I construct a novel and relatively comprehensive data set for the cross-sectional anomaly variables.
    • Empirically, I discuss the volatility-managed portfolios in the Chinese stock market.
    • Data and Codes for constructing the Data are available upon request.
  • Estimating Expected Return Function Nonparametrically: Based on BART (2020)
    [PDF] [Data] [Code]
    • Application of Bayesian Additive Regression Tree (BART) in financial market.
  • Understanding Kernels Applied in Econometrics: A Comparison Perspective Bridging Econometrics and Machine-learning(2020)
    • Heuristically demonstrates how kernel techniques widely exploited in nonparametric econometrics are connected with the one established in machine-learning literature through Mercer’s theorem.
    • One simple empirical application is provided along with this draft as well.
  • Jeffreys' Prior Asymptotically and Approximately Maximizes Expected Information(2019)
    • Heuristically demonstrates why Jeffreys' prior is adopted in many cases when the Bayesian techniques are applied.
Working in Progress
  • How is Fund Investment Exposed to Stock-level Characteristics ? Evidence from U.S. Equity Market (2021)
    • This is joint work with Xiaobin Liu and Tao Zeng.
    • Novel construction of fund-level index measuring exposure of equity funds to firm-level anomalies.
    • Novel extension of IPCA by incorporating \(\ell_{1}/\ell_{q}\)-regularization.
    • Application of IPCA and associated machine-learning methods in unraveling driving factors for fund investment.

Recent Posts

MCMC, Spot Volatility, and Strategic Value of Information

Discussing some of my recent thoughts and ideas on connection between MCMC, volatility and strategic value of information in financial market microstructure.

By far which anomalies (characteristics) universe we have to make to our research live in

Discussion about characteristics construction for asset pricing research.

Variational Bayes Dynamic Variable Selection based on Rcpp and Armadillo

Brief disucssion about implementing variational bayes dynamic variable selection (VBDVS) in hybrid R anc C++ codes.

We are chasing perfectness, but we have to accept flaws

Which Anomalies Matter for Portfolio Construction

This is a brief discussion corresponding the the application of ML in recovering annomaly importance.