Chinese / English
Hong Han

Summary of Achievements of Professor Xiaohong Chen


Professor Xiaohong Chen of the Department of Economics at Yale University is the co-winner of the 2017 best Chinese Economist Award sponsored by the Beijing Foundation of Contemporary Economics. Professor Chen, who holds the title of the Malcolm K. Brachman Professor of Economics of Yale University, is a generational leader of the econometrics profession and has made numerous cutting-edge and path-breaking contributions to both econometrics theory and its applications in numerous areas in economic analysis.


Professor Chen received her B.A. (in mathematics) from Wuhan University, China, and her Ph.D. (in economics) from the University of California-San Diego. Prior to joining Yale University as a full professor of economics in 2007, she has held faculty appointments at the University of Chicago, the London School of Economics, and New York University.


Professor Chen is an elected fellow of the Econometric Society since 2007 (the first one among all those who were born and college-educated in China),and has contributed large scores of scholarly articles and chapters to leading journals in economics and statistics. She is also an international fellow of the Centre for Microdata Methods and Practice in London, and a fellow of the Journal of Econometrics. She has been a special term professor at the Guanghua School of Management, Beijing University and at Shanghai University of Finance and Economics. Professor Chen was awarded membership of Thousand Experts Plan B in China.


Her other awards include the Econometric Theory Multa Scripsit Award, the Richard Stone Prize in Applied Econometrics, and the Arnold Zellner Award in Theoretical Econometrics. She has won multiple grant awards from the National Science Foundation. Professor Chen has served as associate editor of many leading journals including Econometrica, Review of Economic Studies, Quantitative Economics, the Journal of Econometrics, Econometric Theory, the Econometrics Journal, and the Journal of Nonparametric Statistics, among others. She has been an invited speaker at numerous professional conferences, and a program committee member of numerous meetings of the Econometric Society and other professional economics organizations.


Professor Chen is the very leader of the area of sieve estimation and inference methods, an approach that offers precise approximations to analyze complex and high dimensional economic model and data. Prior to Professor Chen’s work, the general-purpose toolkit of conventional econometrics analysis is based on the general theory of inference for finite dimensional parametric models. Nonetheless, it is well known that the presumption of a finite dimensional parametric model is too strong for many economic models. The sieve based nonparametric and semiparametric estimation methods developed by Professor Chen effectively alleviates the reliance on ad hoc parametric assumptions, and allows applies economists to focus on empirical insights and guidance from economic theory, instead of on functional form specification.


Being the prime leader of the sieve estimation method and the author of the Handbook of Econometrics chapter on sieve estimation, Professor Chen’s work has greatly influenced the economics profession, is taught to graduate students in all economics programs, and is becoming classic reference for all researchers engaging in both theoretical and empirical econometric analysis.


In addition, Professor Chen also makes prolific contributions to a wide variety of frontier research areas including treatment effect models, measurement error models, model selection methods, specification testing, copula-based models of dependence, and applications of semiparametric estimators to Engel curve models and to asset pricing kernels under weak assumptions about consumer habit persistence. Her coauthors include many top tier researchers from many areas outside of econometrics including among others labor economics, macroeconomics and financial economics. The positive externalities that she has brought to the economics profession and to the general welfare of societies is tremendous and unmeasurable.


In the area of measurement errors, a major concern with many error-ridden economic data sets is the potential correlation between the measurement errors and the underlying true unknown variables of interest, a phenomenon well documented in the empirical literature. Such correlation violates the classical assumptions and invalidates many conventional solutions for measurement error models. The work by Professor Chen and her coauthors provides a solution to this problem in the context of nonlinear moment conditions by using an auxiliary data set that contains information about the conditional distribution of the true variable given the mismeasured variable. This method allows for the combination of the two data sets to obtain a feasible semiparametric estimator of the parameter of interest.


Furthermore, Professor Chen and her collaborators developed a general framework for analyzing efficient estimation for missing data situations that include measurement error models and treatment effect models as special cases. This high level framework not only derives the semiparametric efficiency bound parameters defined through general nonlinear, possibly nonsmooth and overidentified moment conditions, but also semiparametric efficient estimators using both the conditional expectation projection principle and the inverse probability weighting principle, thus unifying divergent approaches from the previous literature.


Professor Chen is also a specialist of statistical inference in incomplete econometrics models, a topic that is concerned with how to conduct inference when the parameter is only set identified, or how to form statistics for testing economic hypothesis and form areas of confidence to cover the true parameter or the identified set of parameters. Set identified econometric models, have played a key role and are becoming increasingly popular in the field of econometrics in recent years. This area of econometrics is motivated by the fact that many economic and econometric models may involve parameters that are not point identified if a researcher is not willing to impose very strong parametric identifying assumptions that might not be grounded in economic theory alone.


These parameters can only be consistently estimated up to a set. For example, many empirical models of strategic oligopolistic entry model in industrial organization admit the possibility of multiple equilibria. If a researcher is not willing to impose strong parametric assumptions on the equilibrium selection mechanism, then the parameters of the static and dynamic profit structures are often only identified up to a set but may not be point-identified. These models are also referred to as incomplete econometric models. In particular, recent work by Professor Chen and her coauthors demonstrates how to make use of Bayesian computation and Monte Carlo Markov Chain techniques to provide asymptotically consistent confidence sets for partially identified parameters.


Most notably, Professor Chen is also one of the earliest contributors to analyze the econometric and statistical properties of neural network models. Neural network models are becoming the underlying driving engine for data mining, learning machine, and artificial intelligence. Even prior to the massive application of industrial applications of neural networks, Professor Chen’s early work has already developed consistency and asymptotic distributional properties of both nonparametric and semiparametric models estimated by neural networks.


Given the vastness of Professor Chen’s scholarly contributions, it is not possible to enumerate all of Professor Chen’s work in this short document. This brief document only serves to illustrate a small portion of Profession Chen’s profound and record-breaking contributions to economics and to the general society. In addition to being a world leader in econometrics and economics, Professor Chen is also the best professional colleague one can ever have, and a great mentor and advisor to many students.



◆please indicate the source if authorized: National Economics Foundation

◆photo:National Economics Foundation