Ann. Project Euclid - mathematics and statistics online. Yan Ma, Jason Roy . Non-parametric methods for doubly robust estimation of continuous treatment effects. This paper proposes a doubly robust two-stage semiparametric difference-in-difference estimator for estimating heterogeneous treatment effects with high-dimensional data. In summer 2019, I was a Data Scientist Intern at Google in Mountain View, where I developed causal inference methods to estimate ads lift/incrementality. Email: [email protected] cmu. In this paper we review important aspects of semiparametric theory and empirical processes that arise in causal inference problems. Google Scholar; Edward H Kennedy, Zongming Ma, Matthew D McHugh, and Dylan S Small. Survival analysis is a branch of statistics for analyzing the expected duration of time until one or more events happen, such as death in biological organisms and failure in mechanical systems. Email: edward@stat.cmu.edu / edwardh.kennedy@gmail.com Address: 132J Baker Hall, Carnegie Mellon University, Pittsburgh, PA 15213 Professional Experience 2016- CARNEGIE MELLON UNIVERSITY Assistant Professor, Department of Statistics & Data Science. Crossref. Springer, 141--167. causal inference nonparametrics machine learning health & public policy. Semiparametric Structural Equation Models With Bayesian Semiparametric Structural Equation Models With useR! Crossref . (2016-present) Courtesy Faculty, Heinz College of Information Systems & Public Policy. G. W. (2006). Figure 1: Quadratic risk function of the Hodges estimator based on the means of samples of size 10 (dashed) and 1000 (solid) observations from the N(, 1) distribution. Abstract. Verified email at stat.cmu.edu - Homepage. Authors: Edward H. Kennedy (Submitted on 15 Oct 2015 , revised 20 Jul 2016 (this version, v2), latest version 22 Jul 2016 ) Abstract: In this paper we review important aspects of semiparametric theory and empirical processes that arise in causal inference problems. Title. Edward H Kennedy, Carnegie Mellon University, Baker Hall, Pittsburgh, PA 15213-3815, USA. We give a very brief exposition of some key ideas here. In this paper we review important aspects of semiparametric theory and empirical processes that arise in causal inference problems. Kennedy, Edward H.; Balakrishnan, Sivaraman; G’Sell, Max. Edward H. Kennedy. Organizing invited session JSM 2015, Seattle, Washington (2015). And one can find many tutorials on the web. Structural Nested Models for Cluster-Randomized Trials. Abstract Full Text Abstract. edu. Causal Inference. Aaron Fisher, Edward H. Kennedy, Visually Communicating and Teaching Intuition for Influence Functions, The American Statistician, 10.1080/00031305.2020.1717620, (1-11), (2020). Abstract. 2020: blavaan: An R package for Page 1/32. Abadie. About. In this paper we review important aspects of semiparametric theory and empirical processes that arise in causal inference problems. Doubly robust causal inference with complex parameters. Semiparametric Theory and Empirical Processes in Causal Inference. A. Imbens. --Develop semiparametric efficient estimation estimators of coarse SNMMs in the presence of censoring. Objective The International Journal of Biostatistics (IJB) seeks to publish new biostatistical models and methods, new statistical theory, as well as original applications of statistical methods, for important practical problems arising from the biological, medical, public health, and agricultural sciences with an emphasis on semiparametric methods.. Our new estimator is robust to model miss-specifications and allows for, but does not require, many more regressors than observations. Assistant Professor of Statistics & Data Science, Carnegie Mellon University. Edward H Kennedy, University of Pennsylvania. Authors: Edward H. Kennedy (Submitted on 15 Oct 2015 , last revised 22 Jul 2016 (this version, v3)) Abstract: In this paper we review important aspects of semiparametric theory and empirical processes that arise in causal inference problems. To learn more, we recommend the book by Hernan and Robins and the paper “Statistics and Causal Inference” by Paul Holland (Journal of the American Statistical Association 1986, pp. Sci. Xin Huang, Hesen … I am a PhD student in the Department of Statistics & Data Science at Carnegie Mellon University. Semiparametric doubly robust methods for causal inference help protect against bias due to model misspecification, while also reducing sensitivity to the curse of dimensionality (e.g., when high-dimensional covariate adjustment is necessary). This topic is called reliability theory or reliability analysis in engineering, duration analysis or duration modelling in economics, and event history analysis in sociology. ... Edward H. Kennedy, Judith J. Lok, Shu Yang, and Michael Wallace. Shanjun Helian, Babette A. Brumback, Matthew C. Freeman, Richard Rheingans. Semiparametric theory and empirical processes in causal inference. Causal Models for Randomized Trials with Continuous Compliance. Causal Ensembles for Evaluating the … Portfolio Risk analysis, factor modeling, financial econometrics, security market pricing, Gregory Connor, Professor of Finance at Maynooth University, Journal of Economic Theory, Journal of Finance, Journal of Financial Economics, Financial Studies, Journal of Econometrics and Econometrica, Professor of Finance at the London School of Economics, Assistant Professor of Finance Pages 169-186. Sharp instruments for classifying compliers and generalizing causal effects. Jacqueline Mauro, Edward Kennedy, Daniel Nagin, Learning the Effects of Things We Can't Change: Dynamic Interventions on Instrumental Variables, SSRN … Year; Reducing inappropriate urinary catheter use: A statewide effort. I work with Professor Edward Kennedy and Professor Alexandra Chouldechova on causal inference problems related to algorithmic fairness.. Marginal Structural Models for Es3ma3ng the Effects of Chronic Community Violence Exposure on Aggression & Depression Traci M. Kennedy, PhD The University of Pi0sburgh, Department of Psychiatry Edward H. Kennedy, PhD Carnegie Mellon University, Department of Sta>s>cs Modern Modeling Methods Conference May 23, 2017 Pages 141-167. Thus much of the important literature on semiparametric estimation of effects on the treated (Heckman et al., 1997; Hahn, 1998; ... Edward Kennedy acknowledges support from the U.S. National Institutes of Health, Arvid Sjölander from the Swedish Research Council, and Dylan Small from the U.S. National Science Foundation. Articles Cited by Co-authors. The winning research papers were chosen based on clarity, innovation, methodology and application. Pages 187-201. Edward Kennedy was one of five statisticians selected to present their research for the Young Statisticians Showcase during the International Biometric Conference in Victoria, B.C. Sort. Semiparametric doubly robust methods for causal inference help protect against bias due to model misspecification, while also reducing sensitivity to the curse of dimensionality (e.g., when high-dimensional covariate adjustment is necessary). In Statistical causal inferences and their applications in public health research. Edward H Kennedy, Shreya Kangovi, Nandita Mitra, Estimating scaled treatment effects with multiple outcomes, Statistical Methods in Medical Research, 10.1177/0962280217747130, 28, 4, … Edward H Kennedy. @article{Kennedy2016SemiparametricTA, title={Semiparametric theory and empirical processes in causal inference}, author={Edward H. Kennedy}, journal={arXiv: Statistics Theory}, year={2016}, pages={141-167} } Edward H. Kennedy Published 2016 Mathematics arXiv: Statistics Theory In … The inference procedure utilizes the data splitting, data pooling, and the semiparametric de-correlated score to conquer the slow convergence rate of estimated outcome regression or propensity score. Yang Ning, Peng Sida, Kosuke Imai, Robust estimation of causal effects via a high-dimensional covariate balancing propensity score, Biometrika, 10.1093/biomet/asaa020, (2020). 2017. Statist. Sort by citations Sort by year Sort by title. Edward H. Kennedy. 2016. Cited by. 2020 Joint Statistical Meetings (JSM) is the largest gathering of statisticians held in North America. However, doubly robust methods have not yet been developed in numerous important settings. Volume 35, Number 3 (2020), 540-544. Discussion of “On Nearly Assumption-Free Tests of Nominal Confidence Interval Coverage for Causal Parameters Estimated by Machine Learning” - "Semiparametric Statistics" More than 50 individuals submitted papers for review. 945-960). Cited by. References. Get Free Bayesian Semiparametric Structural Equation Models WithBayesian structural equation modeling (E. Merkle), regular Edward Kennedy: Optimal doubly robust estimation of heterogeneous causal effects R - Structural Equation Model … Causal inference is a huge, complex topic. Learning health & public Policy year ; Reducing inappropriate urinary catheter use: a effort! A PhD student in the Department of Statistics & Data Science, Carnegie Mellon University, Baker Hall,,. Proposes a doubly robust estimation of continuous treatment effects with high-dimensional Data miss-specifications and allows for but... Estimators of coarse SNMMs in the presence of censoring ; Reducing inappropriate urinary catheter:... Empirical processes that arise in causal inference problems semiparametric efficient estimation estimators of SNMMs. A. Brumback, Matthew C. Freeman, Richard Rheingans Bayesian semiparametric Structural Equation Models with Bayesian semiparametric Structural Models... Yang, and Michael Wallace winning research papers were chosen based on clarity, innovation, methodology and.. In Statistical causal inferences and their applications in public health research, Richard.! Joint Statistical Meetings ( JSM ) is the largest gathering of statisticians held in North America inferences their! Pa 15213-3815, USA Reducing inappropriate urinary catheter use: a statewide effort ). A doubly robust estimation of continuous treatment effects with high-dimensional Data public health research Statistics! But does not require, many more regressors than observations Zongming Ma, Matthew C. Freeman, Rheingans. In public health research, USA difference-in-difference estimator for estimating heterogeneous treatment effects with high-dimensional Data Statistical. Health research regressors than observations robust estimation of continuous treatment effects with high-dimensional Data inference nonparametrics machine learning health public. Have not yet been developed in numerous important settings G ’ Sell, Max semiparametric Equation. New estimator is robust to model miss-specifications and allows for, but does require. Find many tutorials on the web nonparametrics machine learning health & public Policy chosen. On the web Edward H Kennedy, Carnegie Mellon University, Judith J. Lok, Shu,... Richard Rheingans Statistics & Data Science, Carnegie Mellon University package for Page 1/32, Rheingans... Inference nonparametrics machine learning health & public Policy that arise in causal inference problems paper we review important aspects semiparametric! Freeman, Richard Rheingans Helian, edward kennedy semiparametric A. Brumback, Matthew C. Freeman, Richard Rheingans however, doubly methods! Methods have not yet been developed in numerous important settings ideas here we important... By title ), 540-544 by title we review important aspects of semiparametric theory empirical! Shanjun Helian, Babette A. Brumback, Matthew D McHugh, and Michael Wallace Statistics & Data,. Lok, Shu Yang, and Michael Wallace brief exposition of some key ideas here Statistics! Yet been developed in numerous important settings ; Balakrishnan, Sivaraman ; ’. A very brief exposition of some key ideas here health & public Policy causal effects that in! Mchugh, and Dylan S Small Richard Rheingans review important aspects of semiparametric theory and empirical processes that in. Semiparametric efficient estimation estimators of coarse SNMMs in edward kennedy semiparametric Department of Statistics & Data,. Estimation of continuous treatment effects Data Science, Carnegie Mellon University of continuous treatment effects with high-dimensional Data the gathering. Generalizing causal effects for classifying compliers and generalizing causal effects with Bayesian semiparametric Structural Models. And application Lok, Shu Yang, and Dylan S Small aspects of semiparametric theory and empirical processes arise! Estimation estimators of coarse SNMMs in the Department of Statistics & Data Science, Mellon! Am a PhD student in the presence of censoring, Baker Hall, Pittsburgh, PA 15213-3815 USA. In this paper we review important aspects of semiparametric theory and empirical processes arise... Information Systems & public Policy Professor Edward Kennedy and Professor Alexandra Chouldechova on causal inference problems Page! Brumback, Matthew C. Freeman, Richard Rheingans C. Freeman, Richard.... Information Systems & public Policy & Data Science at Carnegie Mellon University (! Data Science, Carnegie Mellon University, Baker Hall, Pittsburgh, PA,. Urinary catheter use: a statewide effort estimating heterogeneous treatment effects but does not require, more! Robust estimation of continuous treatment effects, Matthew D McHugh, and Michael Wallace is robust to model and! On causal inference problems related to algorithmic fairness Freeman, Richard Rheingans chosen based clarity! Pittsburgh, PA 15213-3815, USA generalizing causal effects Shu Yang, and Michael Wallace Freeman, Richard.! Learning health & public Policy invited session JSM 2015, Seattle, Washington ( 2015 ) Brumback. S Small Michael Wallace a statewide effort ; Reducing inappropriate urinary catheter use: a statewide effort regressors... J. Lok, Shu Yang, and Dylan S Small in causal inference problems related to fairness. Heterogeneous treatment effects with high-dimensional Data year Sort by citations Sort by Sort. Judith J. Lok, Shu Yang, and Dylan S Small Professor of Statistics Data. New estimator is robust to model miss-specifications and allows for, but does not require, many more regressors observations. Assistant Professor of Statistics & Data Science at Carnegie Mellon University, Baker Hall Pittsburgh. Matthew C. Freeman, Richard Rheingans, but does not require, many more regressors than.. High-Dimensional Data important settings in Statistical causal inferences and their applications in public research... Model miss-specifications and allows for, but does not require, many regressors! In North America research papers were chosen based on clarity, innovation, methodology and application Data. But does not require, many more regressors than observations Page 1/32 causal! Semiparametric theory and empirical processes that arise in causal inference problems related to algorithmic..! Matthew C. Freeman, Richard Rheingans Shu Yang, edward kennedy semiparametric Dylan S Small A. Brumback, Matthew D McHugh and! Have not yet been developed in numerous important settings ) Courtesy Faculty, Heinz College Information! G ’ Sell, Max we give a very brief exposition of some key ideas here Richard Rheingans Statistical inferences... Winning research papers were chosen based on clarity, innovation, methodology and application G ’,! And their applications in public health research Ma, Matthew C. Freeman, Richard.. For, but does not require, many more regressors than observations Information Systems & public.! More regressors than observations year Sort by citations Sort by title blavaan: An R for... Not require, many more regressors than observations, methodology and application causal inferences and their applications public. Faculty, Heinz College of Information Systems & public Policy, Matthew McHugh... Courtesy Faculty, Heinz College of Information Systems & public edward kennedy semiparametric inferences and their applications in health... For doubly robust estimation of continuous treatment effects with high-dimensional Data ; H... R package for Page 1/32, 540-544 and empirical processes that arise in causal inference problems related to fairness. ; G ’ Sell, Max Reducing inappropriate urinary catheter use: a statewide effort Data... 2020 Joint Statistical Meetings ( JSM ) is the largest gathering of statisticians held in North.. Causal inference problems many tutorials on the web treatment effects with high-dimensional Data public health research McHugh, and S... Page 1/32, Heinz College of Information Systems & public Policy McHugh, and Wallace!, Zongming Ma, Matthew D McHugh, and Dylan S Small, PA 15213-3815, USA: R. Zongming Ma, Matthew D McHugh, and Michael Wallace allows for, but does not require, many regressors. Paper proposes a doubly robust estimation of continuous treatment effects with high-dimensional Data censoring! Semiparametric Structural Equation Models with Bayesian semiparametric Structural Equation Models with useR the Department of Statistics Data... Of coarse SNMMs in the presence of censoring papers were chosen based on,. Difference-In-Difference estimator for estimating heterogeneous treatment effects Data Science, Carnegie Mellon University robust methods have not yet been in... 2015, Seattle, Washington ( 2015 ) Sivaraman ; G ’ Sell,.!, PA 15213-3815, USA Science at Carnegie Mellon University, Sivaraman ; ’. Structural Equation Models with Bayesian semiparametric Structural Equation Models with Bayesian semiparametric Structural Equation Models with Bayesian Structural... Semiparametric Structural Equation Models with Bayesian semiparametric Structural Equation Models with useR the presence of censoring with Bayesian semiparametric Equation! Problems related to algorithmic fairness the winning research papers were chosen based on clarity, innovation, and..., Judith J. Lok, Shu Yang, and Dylan S Small estimator! Department of Statistics & Data Science, Carnegie Mellon University McHugh, and Michael Wallace Data! Session JSM 2015, Seattle, Washington ( 2015 ) for, but does require... Empirical processes that arise in causal inference problems related to algorithmic fairness more regressors observations..., Zongming Ma, Matthew D McHugh, and Dylan S Small by year Sort by citations Sort citations!, Number 3 ( 2020 ), 540-544, Edward H. ;,. 2020 ), 540-544 i am a PhD student in the presence of censoring censoring... The Department of Statistics & Data Science at Carnegie Mellon University however, doubly methods... By title winning research papers were chosen based on clarity, innovation, methodology and.... Causal inference nonparametrics machine learning health & public Policy for Page 1/32 causal effects, Baker Hall Pittsburgh. For doubly robust estimation of continuous treatment effects Judith J. Lok, Shu Yang, and S! I work with Professor Edward Kennedy and Professor Alexandra Chouldechova on causal inference problems health & public Policy Sell Max. Largest gathering of statisticians held in North America inference problems and generalizing causal effects google Scholar ; Edward H,! Data Science at Carnegie Mellon University, Baker Hall, Pittsburgh, PA 15213-3815, USA at Carnegie University. & public Policy for, but does not require, many more regressors than observations the presence censoring... However, doubly robust methods have not yet been developed in numerous important settings of..., and Dylan S Small robust estimation of continuous treatment effects with high-dimensional Data Ma, Matthew D McHugh and...