Dacheng Xiu's Homepage

Biography

DACHENG XIU

Professor of Econometrics and Statistics
Booth School of Business, University of Chicago

5807 South Woodlawn Avenue, Chicago, IL 60637
773.834.7191
dacheng.xiu@chicagobooth.edu
Google Scholar Profile
Curriculum Vitae

Dacheng Xiu’s research interests include developing statistical methodologies and applying them to financial data, while exploring their economic implications. His earlier research involved risk measurement and portfolio management with high-frequency data and econometric modeling of derivatives. His current work focuses on developing machine learning solutions to big-data problems in empirical asset pricing.

Xiu’s work has appeared in Econometrica, Journal of Econometrics, Journal of the American Statistical Association, Annals of Statistics, Review of Financial Studies, and Journal of Finance. He is a Co-Editor for the Journal of Financial Econometrics, an Associate Editor for the Journal of Econometrics, Journal of Business & Economic Statistics, Management Science, Journal of Applied Econometrics, Journal of Empirical Finance, and Statistica Sinica. He has received several recognitions for his research, including the Fellow of the Society for Financial Econometrics, the Fellow of the Journal of Econometrics, the 2018 Swiss Finance Institute Outstanding Paper Award, the 2018 AQR Insight Award, and the Best Conference Paper Prize at the 2017 Annual Meeting of the European Finance Association. In 2017, Xiu launched a website that provides up-to-date realized volatilities of individual stocks, as well as equity, currency, and commodity futures. These daily volatilities are calculated from intraday transactions and the methodologies are based on his research of high-frequency data.

Xiu earned his PhD and MA in applied mathematics from Princeton University, where he was also a student at the Bendheim Center for Finance. Prior to his graduate studies, he obtained a BS in mathematics from the University of Science and Technology of China.

Research

PUBLICATIONS

"Autoencoder Asset Pricing Models", with Shihao Gu and Bryan Kelly, forthcoming in the Journal of Econometrics.

Machine Learning Time-Series and Cross-Section of Expected Returns Nonlinear Factor Model Neural Networks Big Data Return Predictability

"High-Frequency Factor Models and Regressions", with Yacine Aït-Sahalia and Ilze Kalnina, forthcoming in the Journal of Econometrics.

High Frequency Fama-French Factors Volatility Estimation Factor Model

"Empirical Asset Pricing via Machine Learning", with Shihao Gu and Bryan Kelly, forthcoming in the Review of Financial Studies. Winner of the 2018 Swiss Finance Institute Outstanding Paper Award.

Chicago Booth Review GitHub Empirical Data Machine Learning Time-Series and Cross-Section of Expected Returns Neural Networks Big Data Return Predictability

"Taming the Factor Zoo: A Test of New Factors", with Guanhao Feng and Stefano Giglio, forthcoming in the Journal of Finance. First Prize Winner of the 2018 AQR Insight Award.

Chicago Booth Review Internet Appendix.pdf Slides Machine Learning Model Selection Cross-Section of Expected Returns Big Data

"A Hausman Test for the Presence of Market Microstructure Noise in High Frequency Data", with Yacine Aït-Sahalia, Journal of Econometrics 211 (2019), 176-205.

Matlab Codes.zip Microstructure Noise QMLE Volatility Estimation

"Knowing Factors or Factor Loadings, or Neither? Evaluating Estimators of Large Covariance Matrices with Noisy and Asynchronous Data", with Chaoxing Dai and Kun Lu, Journal of Econometrics 208 (2019), 43-79.

Factor Model Covariance Estimation Microstructure Noise MSCI Barra Model PCA Portfolio Allocation Big Data

"Efficient Estimation of Integrated Volatility Functionals via Multiscale Jackknife", with Jia Li and Yunxiao Liu, Annals of Statistics Vol. 47, No. 1, (2019), 156-176.

Spot (Co)Variance Jackknife Bootstrap

"Principal Component Analysis of High Frequency Data", with Yacine Aït-Sahalia, Journal of the American Statistical Association 114 (2019), 287-303.

Biplots.mp4 PCA Spot (Co)Variance

"Resolution of Policy Uncertainty and Sudden Declines in Volatility", with Dante Amengual, Journal of Econometrics 203 (2018), 297-315.

Chicago Booth Review Supplemental Material.pdf Variance Swaps Pricing Non-affine Models Downward Volatility Jumps VIX

"Econometric Analysis of Multivariate Realised QML: Efficient Positive Semi-definite Estimators of the Covariation of Equity Prices", with Neil Shephard, Journal of Econometrics 201 (2017), 19-42.

Matlab Codes.zip QMLE Factor Model Covariance Estimation Microstructure Noise Kalman Filtering, Smoothing, and EM Algorithm

"Nonparametric Estimation of the Leverage Effect: A Trade-off between Robustness and Efficiency", with Ilze Kalnina, Journal of the American Statistical Association 112 (2017), 384-396.

Covariance Estimation VIX Spot (Co)Variance

"Using Principal Component Analysis to Estimate a High Dimensional Factor Model with High-Frequency Data", with Yacine Aït-Sahalia, Journal of Econometrics 201 (2017), 384-399.

High Frequency Fama-French Factors Big Data Factor Model PCA Portfolio Allocation Covariance Estimation

"Generalized Methods of Integrated Moments for High-Frequency Data", with Jia Li, Econometrica 84 (2016), 1613-1633.

Chicago Booth Review Supplemental Material.pdf An earlier version: Spot Variance Regressions.pdf High Frequency GMM Spot (Co)Variance

"Increased Correlation Among Asset Classes: Are Volatility or Jumps to Blame, or Both?" , with Yacine Aït-Sahalia, Journal of Econometrics 194 (2016), 205-219.

Co-Jumps Covariance Estimation Microstructure Noise

"Incorporating Global Industrial Classification Standard into Portfolio Allocation: A Simple Factor-Based Large Covariance Matrix Estimator with High Frequency Data", with Jianqing Fan and Alex Furger, Journal of Business & Economic Statistics 34 (2016), 489-503, (Special Issue on Big Data).

Chicago Booth Review Matlab Codes.zip High Frequency Fama-French Factors Portfolio Allocation Big Data Covariance Estimation

"A Tale of Two Option Markets: Pricing Kernels and Volatility Risk", with Zhaogang Song, Journal of Econometrics 190 (2016), 176-196. Honorable mention of the 2017 Dennis J. Aigner Award.

Nonparametric Option Pricing Closed-form Pricing of SPX and VIX Options Particle Filtering

"Quasi-Maximum Likelihood Estimation of GARCH Models with Heavy-Tailed Likelihoods", with Jianqing Fan and Lei Qi, Journal of Business & Economic Statistics 32 (2014), 178-191. Invited Paper with Discussion.

Rejoinder.pdf Matlab Codes.zip Volatility Estimation

"Hermite Polynomial Based Expansion of European Option Prices", Journal of Econometrics 179 (2014), 158-177.

Option Pricing Non-affine Models

"High-Frequency Covariance Estimates with Noisy and Asynchronous Data", with Yacine Aït-Sahalia and Jianqing Fan, Journal of the American Statistical Association 105 (2010), 1504-1517.

Matlab Codes.zip QMLE Covariance Estimation Microstructure Noise

"Quasi-Maximum Likelihood Estimation of Volatility with High Frequency Data", Journal of Econometrics 159 (2010), 235-250.

QMLE Volatility Estimation Microstructure Noise

WORKING PAPERS

"When Moving-Average Models Meet High-Frequency Data: Uniform Inference on Volatility", with Rui Da, May 2019. Submitted.

Chicago Booth Review Supplemental Material QMLE Volatility Estimation Microstructure Noise Kalman Filtering Model Selection

"Asset Pricing with Omitted Factors", with Stefano Giglio, Sep. 2019. Submitted. Winner of the Best Conference Paper Prize at the 44th EFA.

Chicago Booth Review Slides Matlab Codes.zip Three-Pass Estimator Factor Model PCA Cross-Section of Expected Returns Big Data

"Thousands of Alpha Tests", with Stefano Giglio and Yuan Liao, Mar. 2019. In Revision.

Data Snooping Hedge Funds Factor Model PCA Cross-Section of Expected Returns Machine Learning Big Data

"Predicting Returns with Text Data", with Tracy Ke and Bryan Kelly, Jul. 2019. Submitted. Winner of the 2019 CICF Best Paper Award.

Text Mining Machine Learning Cross-Section of Expected Returns Big Data Return Predictability Sentiment Analysis Topic Model

"The Structure of Economic News", with Leland Bybee, Bryan Kelly, and Asaf Manela, Sep. 2019.

The Structure of Economic News Website Text Mining Machine Learning Big Data Topic Model

COMMENTS & BOOK CHAPTERS

"Comment on: Limit of Random Measures Associated with the Increments of a Brownian Semimartingale", with Jia Li, Journal of Financial Econometrics 16(4) (2018), 570-582.

"Likelihood-Based Volatility Estimators in the Presence of Market Microstructure Noise: A Review", with Yacine Aït-Sahalia, Handbook of Volatility Models and their Applications, Chapter 14, Wiley 2012.

ETFs	12/03/2019	12/02/2019	11/29/2019	11/27/2019	11/26/2019
0Market S&P500SPY	09.75% ±0.4%	09.85% ±0.5%	04.05% ±0.2%	03.76% ±0.1%	04.67% ±0.1%
MaterialXLB	11.69% ±1.4%	13.17% ±1.9%	08.12% ±0.6%	08.20% ±0.7%	07.25% ±0.9%
IndustrialXLI	12.49% ±1.4%	12.81% ±1.7%	06.06% ±0.9%	06.45% ±0.3%	06.51% ±0.6%
Consumer DiscretionaryXLY	09.60% ±1.3%	09.37% ±1.5%	05.76% ±3.9%	05.27% ±0.7%	04.70% ±0.9%
Consumer StaplesXLP	08.77% ±0.7%	08.95% ±0.5%	05.58% ±0.7%	05.83% ±0.6%	05.73% ±0.6%
Health CareXLV	10.23% ±1.2%	11.58% ±2.4%	07.34% ±0.6%	06.73% ±0.7%	06.95% ±0.5%
FinancialXLF	11.61% ±0.8%	11.45% ±1.0%	05.91% ±0.6%	05.92% ±0.3%	06.98% ±0.4%
TechnologyXLK	13.06% ±1.6%	13.55% ±1.9%	05.09% ±0.5%	05.43% ±0.3%	05.99% ±0.6%
UtilitiesXLU	08.90% ±0.5%	10.14% ±0.9%	09.09% ±0.9%	08.93% ±0.8%	07.82% ±0.6%
EnergyXLE	17.87% ±0.7%	16.60% ±1.1%	11.40% ±0.7%	12.22% ±1.0%	12.36% ±0.9%

Objective
We provide up-to-date daily annualized realized volatilities for individual stocks, ETFs, and future contracts, which are estimated from high-frequency data. We are in the process of incorporating equities from global markets.

Data
We collect trades at their highest frequencies available (up to every millisecond for US equities after 2007), and clean them using the prevalent national best bid and offer (NBBO) that are available up to every second. The mid-quotes are calculated based on the NBBOs, so their highest sampling frequencies are also up to every second.

Methodology
We provide quasi-maximum likelihood estimates of volatility (QMLE) based on moving-average models MA(q), using non-zero returns of transaction prices (or mid-quotes if available) sampled up to their highest frequency available, for days with at least 12 observations. We select the best model (q) using Akaike Information Criterion (AIC). For comparison, we report realized volatility (RV) estimates using 5-minute and 15-minute subsampled returns.

References
1. “When Moving-Average Models Meet High-Frequency Data: Uniform Inference on Volatility”, by Rui Da and Dacheng Xiu. 2017.
2. “Quasi-Maximum Likelihood Estimation of Volatility with High Frequency Data”, by Dacheng Xiu. Journal of Econometrics, 159 (2010), 235-250.
3. “How Often to Sample A Continuous-time Process in the Presence of Market Microstructure Noise”, by Yacine Aït-Sahalia, Per Mykland, and Lan Zhang. Review of Financial Studies, 18 (2005), 351–416.
4. “The Distribution of Exchange Rate Volatility”, by Torben Andersen, Tim Bollerslev, Francis X. Diebold, and Paul Labys. Journal of the American Statistical Association, 96 (2001), 42-55.
5. “Econometric Analysis of Realized Volatility and Its Use in Estimating Stochastic Volatility Models”, by Ole E Barndorff‐Nielsen and Neil Shephard. Journal of the Royal Statistical Society: Series B, 64 (2002), 253-280.

Teaching

B41100 Applied Regression Analysis (MBA)

This course is about regression, a powerful and widely used data analysis technique wherein we seek to understand how different random quantities relate to one another. Students will learn how to use regression to analyze a variety of complex real world problems, with the aim of understanding data and prediction of future events. Focus is placed on understanding of fundamental concepts and development of the skills necessary for robust application of regression techniques. Examples are used throughout to illustrate application of the tools.

B41902 Statistical Inference (PhD)

The focus of this course will be methods to draw inferences in econometric models. We will cover linear regression models, GMM, nonlinear models, and time series models. The majority of the discussion will cover frequentist methods focusing on the use of approximations to finite-sample sampling distributions as a means for obtaining inference. We will cover methods that are appropriate for independent data as well as dependent data. We will discuss intuition for how and when to use the econometric tools developed in the class in addition to deriving some of the relevant theoretical properties.

2018 SoFiE Summer School

Machine Learning and Finance: The New Empirical Asset Pricing. Program is here.