Tetsuya Kaji
I am Associate Professor of Econometrics and Statistics at the University of Chicago Booth School of Business. I work at the intersections of economics, statistics, and machine learning.
Class Material
BUSN 41901 Probability and Statistics
This is a PhD course that introduces fundamental statistical concepts for academic research in business and economics. [Lecture Notes]
Working Papers
Necessary and Sufficient Conditions for Convergence in Distribution of Quantile and P-P Processes in L1(0,1)
The necessary and sufficient condition for the empirical quantile process to converge in L1 is given. In addition to the necessary and sufficient condition for the empirical cdf process to converge in L1, it requires local absolute continuity of the quantile function. (with Brendan Beare)
Assessing Heterogeneity of Treatment Effects
We derive sharp bounds on the subgroup treatment effects and the subgroup proportions of winners, which complement the quantile treatment effects. (with Jianfei Cao)
Controlling Tail Risk Measures with Estimation Error
Risk control with Value-at-Risk and expected shortfall bounds the probability of bad events. We construct a way to bound the “true probability” when a valid confidence interval is available for the risk measures. (with Hyungjune Kang)
Asymptotic Theory of L-Statistics and Integrable Empirical Processes
The main result of this paper has been generalized and is contained in “Necessary and Sufficient Conditions for Convergence in Distribution of Quantile and P-P Processes in L1(0,1)” with Brendan Beare.
Publications
An Adversarial Approach to Structural Estimation
Econometrica, 91(6), 2041−2063, November 2023 (with Elena Manresa and Guillaume Pouliot). [Preprint] [Online Appendix] [Code] [Chicago Booth Review]
Metropolis-Hastings via Classification
Journal of the American Statistical Association, 118(544), 2533−2547, 2023 (with Veronika Ročková). [Preprint] [Online Appendix]
Approximate Bayesian Computation via Classification
Journal of Machine Learning Research, 23(350), 1−49, 2022 (with Yuexi Wang and Veronika Ročková). [Preprint]
Adversarial Inference Is Efficient
AEA Papers and Proceedings, 111, 621−625, May 2021 (with Elena Manresa and Guillaume Pouliot). [Online Appendix]
Theory of Weak Identification in Semiparametric Models
Econometrica, 89(2), 733−763, March 2021. [Preprint] [Code]
Extremal Quantile Regression
Handbook of Quantile Regression, ed. by R. Koenker, V. Chernozhukov, X. He, and L. Peng, Chapman & Hall/CRC, 2017, ch. 18, 333−362 (with Victor Chernozhukov and Iván Fernández-Val). [Preprint] [Code]