About

Donald W. K. Andrews is the T. C. Koopmans Professor of Economics at Yale University, where he also serves as a Professor of Statistics and Data Science. He is an elected fellow of the Econometric Society and the American Academy of Arts and Sciences, and a founding fellow of the International Association of Applied Econometricians. Andrews has received multiple awards, including the Plura Scripsit and Plurima Scripsit Econometric Theory Awards, as well as numerous teacher and advisor of the year honors. His research specializes in econometric theory, with interests encompassing inference under partial and weak identification, uniformity in asymptotic approximations, time series analysis, structural change testing, bootstrap methods, semiparametric and nonparametric estimation, empirical process theory, computational methods, and robust estimation and testing.

Research topics

• Computer Science
• Artificial Intelligence
• Sociology
• Mathematics
• Philosophy
• Psychology
• Internet privacy
• Medicine
• Operations research
• Engineering
• Mathematical economics
• Law and economics
• Business
• Risk analysis (engineering)
• Management science

Selected publications

• Initial-Condition-Robust Inference in Autoregressive Models

  ArXiv.org · 2026-02-10

  articleOpen access1st authorCorresponding

  This paper considers confidence intervals (CIs) for the autoregressive (AR) parameter in an AR model with an AR parameter that may be close or equal to one. Existing CIs rely on the assumption of a stationary or fixed initial condition to obtain correct asymptotic coverage and good finite sample coverage. When this assumption fails, their coverage can be quite poor. In this paper, we introduce a new CI for the AR parameter whose coverage probability is completely robust to the initial condition, both asymptotically and in finite samples. This CI pays only a small price in terms of its length when the initial condition is stationary or fixed. The new CI also is robust to conditional heteroskedasticity of the errors.

  PublisherOA PDF

• Initial-Condition-Robust Inference in Autoregressive Models

  Open MIND · 2026-02-10

  preprint1st authorCorresponding

  This paper considers confidence intervals (CIs) for the autoregressive (AR) parameter in an AR model with an AR parameter that may be close or equal to one. Existing CIs rely on the assumption of a stationary or fixed initial condition to obtain correct asymptotic coverage and good finite sample coverage. When this assumption fails, their coverage can be quite poor. In this paper, we introduce a new CI for the AR parameter whose coverage probability is completely robust to the initial condition, both asymptotically and in finite samples. This CI pays only a small price in terms of its length when the initial condition is stationary or fixed. The new CI also is robust to conditional heteroskedasticity of the errors.

  DOI

• Initial-Condition-Robust Inference in Autoregressive Models

  SSRN Electronic Journal · 2026-01-01

  preprintOpen access
  PublisherDOI

• Inference in a stationary/nonstationary autoregressive time‐varying‐parameter model

  Quantitative Economics · 2025-01-01 · 1 citations

  articleOpen access1st authorCorresponding

  This paper considers nonparametric estimation and inference in first‐order autoregressive (AR(1)) models with deterministically time‐varying parameters. A key feature of the proposed approach is to allow for time‐varying stationarity in some time periods, time‐varying nonstationarity (i.e., unit root or local‐to‐unit root behavior) in other periods, and smooth transitions between the two. The estimation of the AR parameter at any time point is based on a local least squares regression method, where the relevant initial condition is endogenous. We obtain limit distributions for the AR parameter estimator and t‐statistic at a given point τ in time when the parameter exhibits unit root, local‐to‐unity, or stationary/stationary‐like behavior at time τ . These results are used to construct confidence intervals and median‐unbiased interval estimators for the AR parameter at any specified point in time. The confidence intervals have correct asymptotic coverage probabilities with the coverage holding uniformly over stationary and nonstationary behavior of the observations.

  PublisherDOI

• Inference in a Stationary/Nonstationary Autoregressive Time-Varying-Parameter Model

  arXiv (Cornell University) · 2024-11-01

  preprintOpen access1st authorCorresponding

  This paper considers nonparametric estimation and inference in first-order autoregressive (AR(1)) models with deterministically time-varying parameters. A key feature of the proposed approach is to allow for time-varying stationarity in some time periods, time-varying nonstationarity (i.e., unit root or local-to-unit root behavior) in other periods, and smooth transitions between the two. The estimation of the AR parameter at any time point is based on a local least squares regression method, where the relevant initial condition is endogenous. We obtain limit distributions for the AR parameter estimator and t-statistic at a given point $τ$ in time when the parameter exhibits unit root, local-to-unity, or stationary/stationary-like behavior at time $τ$. These results are used to construct confidence intervals and median-unbiased interval estimators for the AR parameter at any specified point in time. The confidence intervals have correct asymptotic coverage probabilities with the coverage holding uniformly over stationary and nonstationary behavior of the observations.

  PublisherOA PDFDOI

• A simulation-deep reinforcement learning (SiRL) approach for epidemic control optimization

  Annals of Operations Research · 2022 · 40 citations

  • Computer Science
  • Artificial Intelligence
  • Computer Science
  PublisherOA PDFDOI

• GUEST EDITORS’ INTRODUCTION PART ONE: SPECIAL DUAL ISSUE OF <i>ECONOMETRIC THEORY</i> ON YALE 2018 CONFERENCE IN HONOR OF PETER C. B. PHILLIPS

  Econometric Theory · 2022

  1st authorCorresponding
  • Computer Science
  • Sociology
  • Mathematical economics

  An abstract is not available for this content so a preview has been provided. Please use the Get access link above for information on how to access this content.

  PublisherDOI

• Generic results for establishing the asymptotic size of confidence sets and tests

  Journal of Econometrics · 2020-05-17

  preprint1st authorCorresponding
  PublisherDOI

• Identification- and Singularity-Robust Inference for Moment Condition Models

  SSRN Electronic Journal · 2019-01-01 · 4 citations

  articleOpen access1st authorCorresponding
  PublisherDOI

• Inference in Moment Inequality Models That Is Robust to Spurious Precision under Model Misspecification

  SSRN Electronic Journal · 2019-01-01 · 10 citations

  articleOpen access1st authorCorresponding
  PublisherDOI

Recent grants

• Robust Inference in Econometrics

  NSF · $226k · 2017–2021

• Advances in Econometrics for Treatment Effect Bounds, Time-Varying-Parameter Nonstationary/Stationary Autoregressive Models, and Identification-Robust Inference

  NSF · $258k · 2014–2019

• Adaptive Estimation, the Block-Block Bootstrap, Optimal Tests with Weak Instruments, and Inference with Common Shocks

  NSF · $243k · 2004–2008

• Inference in Econometric Models with Asymptotic Discontinuities

  NSF · $210k · 2008–2012

• Functional Limit Theory in Econometrics

  NSF · $209k · 1992–1995

Frequent coauthors

• James H. Stock

  Harvard University

  35 shared

• Patrik Guggenberger

  Pennsylvania State University

  35 shared

• Marcelo J. Moreira

  15 shared

• Moshe Buchinsky

  University of California, Los Angeles

  14 shared

• Xiaoxia Shi

  14 shared

• Xu Cheng

  14 shared

• Werner Ploberger

  12 shared

• Peter C.B. Phillips

  9 shared

Awards & honors

• Elected fellow of the Econometric Society
• Fellow of the American Academy of Arts and Sciences
• Founding fellow of the International Association of Applied…
• Fellow of the Journal of Econometrics
• Plura Scripsit Award