2016

Double/Debiased Machine Learning for Treatment and Causal Parameters

Chernozhukov, Victor, Chetverikov, Denis, Demirer, Mert et al.

Understand

Most modern supervised statistical/machine learning (ML) methods are explicitly designed to solve prediction problems very well.

Achieving this goal does not imply that these methods automatically deliver good estimators of causal parameters.
Examples of such parameters include individual regression coefficients, average treatment effects, average lifts, and demand or supply elasticities.
In fact, estimates of such causal parameters obtained via naively plugging ML estimators into estimating equations for such parameters can behave very poorly due to the regularization bias.

Built on

Optimal asymptotic tests of composite statistical hypotheses
Neyman, J. (1959) · 1959
Earlier work this paper cites.
Mathematical Statistics: A Decision Theoretic Approach
Ferguson, T. (1967) · 1967
Earlier work this paper cites.
On the efficiency of a class of nonparametric estimates
Levit, B. Y. (1975) · 1975
Earlier work this paper cites.
On one informational lower bound
Nevelson, M. (1977) · 1977
Earlier work this paper cites.
On the nonparametric estimation of functionals
Hasminskii, R. and I. Ibragimov (1978) · 1978
Earlier work this paper cites.
c ⁡ ( α ) c(\alpha) tests and their use
Neyman, J. (1979) · 1979
Earlier work this paper cites.
Statistical Estimation: Asymptotic Theory
Ibragimov, I. A. and R. Z. Hasminskii (1981) · 1981
Earlier work this paper cites.
On adaptive estimation
Bickel, P. J. (1982) · 1982
Earlier work this paper cites.
Large sample properties of generalized method of moments estimators
Hansen, L. (1982) · 1982
Earlier work this paper cites.
The central role of the propensity score in observational studies for causal effects
Rosenbaum, P. R. and D. B. Rubin (1983) · 1983
Earlier work this paper cites.
On asymptotically efficient estimation in semiparametric models
Schick, A. (1986) · 1986
Earlier work this paper cites.
Asymptotic efficiency in estimation with conditional moment restrictions
Chamberlain, G. (1987) · 1987
Earlier work this paper cites.
Estimating integrated squared density derivatives
Bickel, P. and Y. Ritov (1988) · 1988
Earlier work this paper cites.
Root- N N -consistent semiparametric regression
Robinson, P. M. (1988) · 1988
Earlier work this paper cites.
Semiparametric efficiency bounds
Newey, W. (1990) · 1990
Earlier work this paper cites.
On differentiable functionals
van der Vaart, A. W. (1991) · 1991
Earlier work this paper cites.
Specification testing and quasi-maximum-likelihood estimation
Wooldridge, J. (1991) · 1991
Earlier work this paper cites.
Efficiency bounds for semiparametric regression
Chamberlain, G. (1992) · 1992
Earlier work this paper cites.
Profile likelihood and conditionally parametric models
Severini, T. A. and W. H. Wong (1992) · 1992
Earlier work this paper cites.
Identification and estimation of local average treatment effects
Imbens, G. and J. Angrist (1994) · 1994
Earlier work this paper cites.
The asymptotic variance of semiparametric estimators
Newey, W. (1994) · 1994
Earlier work this paper cites.
Split-sample instrumental variables estimates of the return to schooling
Angrist, J. D. and A. B. Krueger (1995) · 1995
Earlier work this paper cites.
Semiparametric efficiency in multivariate regression models with missing data
Robins, J. and A. Rotnitzky (1995) · 1995
Earlier work this paper cites.
Edgeworth approximation for MINPIN estimators in semiparametric regression models
Linton, O. (1996) · 1996
Earlier work this paper cites.
Efficient and Adaptive Estimation for Semiparametric Models
Bickel, P. J., C. A. J. Klaassen, Y. Ritov, and J. A. Wellner (1998) · 1998
Earlier work this paper cites.
On the role of the propensity score in efficient semiparametric estimation of average treatment effects
Hahn, J. (1998) · 1998
Earlier work this paper cites.
Undersmoothing and bias corrected functional estimation
Newey, W. K., F. Hsieh, and J. Robins (1998) · 1998
Earlier work this paper cites.
Asymptotic Statistics
van der Vaart, A. W. (1998) · 1998
Earlier work this paper cites.

Similar

Improved rates and asymptotic normality for nonparametric neural network estimators
Chen, X. and H. White (1999) · 1999
Cited alongside, same era.
Rejoinder to “adjusting for non-ignorable drop-out using semiparametric non-response models”
Scharfstein, D. O., A. Rotnitzky, and J. M. Robins (1999) · 1999
Cited alongside, same era.
Sequential testing of duration data: The case of the Pennsylvania ‘reemployment bonus’ experiment
Bilias, Y. (2000) · 2000
Cited alongside, same era.
The colonial origins of comparative development: An empirical investigation
Acemoglu, D., S. Johnson, and J. A. Robinson (2001) · 2001
Cited alongside, same era.
Quantile regression for duration data: A reappraisal of the pennsylvania reemployment bonus experiments
Bilias, Y. and R. Koenker (2002) · 2002
Cited alongside, same era.
Estimation of semiparametric models when the criterion function is not smooth
Chen, X., O. Linton, and I. van Keilegom (2003) · 2003
Cited alongside, same era.
Efficient estimation of average treatment effects using the estimated propensity score
Hirano, K., G. W. Imbens, and G. Ridder (2003) · 2003
Cited alongside, same era.
The effects of 401 (k) participation on the wealth distribution: an instrumental quantile regression analysis
Chernozhukov, V. and C. Hansen (2004) · 2004
Cited alongside, same era.
Twicing kernels and a small bias property of semiparametric estimators
Newey, W. K., F. Hsieh, and J. M. Robins (2004) · 2004
Cited alongside, same era.
A c ⁡ ( α ) c(\alpha) -type gradient test in the GMM approach
Lee, L. (2005) · 2005
Cited alongside, same era.
Large sample properties of matching estimators for average treatment effects
Abadie, A. and G. W. Imbens (2006) · 2006
Cited alongside, same era.
Targeted maximum likelihood learning
van der Laan, M. and D. Rubin (2006) · 2006
Cited alongside, same era.

Then

Targeted Learning: Causal Inference for Observational and Experimental Data
van der Laan, M. J. and S. Rose (2011) · 2011
Later among the works it cites.
Cross-validated targeted minimum-loss-based estimation
Zheng, W. and M. J. van der Laan (2011) · 2011
Later among the works it cites.
The semiparametric efficiency bound for models of sequential moment restrictions containing unknown functions
Ai, C. and X. Chen (2012) · 2012
Later among the works it cites.
Gaussian approximation of suprema of empirical processes
Chernozhukov, V., D. Chetverikov, and K. Kato (2014) · 2012
Later among the works it cites.
Variance estimation using refitted cross-validation in ultrahigh dimensional regression
Fan, J., S. Guo, and K. Yu (2012) · 2012
Later among the works it cites.
Confidence intervals for low-dimensional parameters with high-dimensional data
Zhang, C. and S. Zhang (2014) · 2012
Later among the works it cites.
Program evaluation with high-dimensional data
Belloni, A., V. Chernozhukov, I. Fernández-Val, and C. Hansen (2017) · 2013
Later among the works it cites.
Uniform post selection inference for lad regression models and other z-estimators
Belloni, A., V. Chernozhukov, and K. Kato (2015) · 2013
Later among the works it cites.
Post-selection inference for generalized linear models with many controls
Belloni, A., V. Chernozhukov, and Y. Wei (2016) · 2013
Later among the works it cites.
Estimating treatment effect heterogeneity in randomized program evaluation
Imai, K. and M. Ratkovic (2013) · 2013
Later among the works it cites.
Hypothesis testing in high-dimensional regression under the gaussian random design model: Asymptotic theory
Javanmard, A. and A. Montanari (2014b) · 2013
Later among the works it cites.
New statistical approaches to semiparametric regression with application to air pollution research
Robins, J., P. Zhang, R. Ayyagari, R. Logan, E. Tchetgen, L. Li, A. Lumley, and A. van der Vaart (2013) · 2013
Later among the works it cites.
On asymptotically optimal confidence regions and tests for high-dimensional models
van de Geer, S., P. Bühlmann, Y. Ritov, and R. Dezeure (2014) · 2013
Later among the works it cites.
Robust inference on average treatment effects with possibly more covariates than observations
Farrell, M. (2015) · 2015
Later among the works it cites.
The influence function of semiparametric estimators
Original
Ichimura, H. and W. Newey (2015) · 2015
Later among the works it cites.
Causal Inference for Statistics, Social, and Biomedical Sciences: An Introduction
Imbens, G. W. and D. B. Rubin (2015) · 2015
Later among the works it cites.
Testing-based forward model selection
Original
Kozbur, D. (2016) · 2015
Later among the works it cites.
A generally efficient targeted minimum loss based estimator
van der Laan, M. J. (2015) · 2015
Later among the works it cites.
Adaptive concentration of regression trees, with application to random forests
Original
Wager, S. and G. Walther (2016) · 2015
Later among the works it cites.
Approximate residual balancing: De-biased inference of average treatment effects in high-dimensions
Original
Athey, S., G. Imbens, and S. Wager (2016) · 2016
Closest in time.
Locally robust semiparametric estimation
Original
Chernozhukov, V., J. Escanciano, H. Ichimura, W. Newey, and J. Robins (2016) · 2016
Closest in time.
Statistical inference for data adaptive target parameters
Hubbard, A. E., S. Kherad-Pajouh, and M. J. van der Laan (2016) · 2016
Closest in time.
Optimal individualized treatments in resource-limited settings
Luedtke, A. R. and M. J. van der Laan (2016) · 2016
Closest in time.
High-dimensional l 2 l_{2} boosting: Rate of convergence
Original
Luo, Y. and M. Spindler (2016) · 2016
Closest in time.
TMLE for marginal structural models based on an instrument
Toth, B. and M. J. van der Laan (2016) · 2016
Closest in time.
Marginal structural models with counterfactual effect modifiers
Zheng, W., Z. Luo, and M. J. van der Laan (2016) · 2016
Closest in time.
Minimax estimation of a functional on a structured high dimensional model
Robins, J., L. Li, R. Mukherjee, E. Tchetgen, and A. van der Vaart (2017) · 2017
Closest in time.
Super learner
van der Laan, M. J., E. C. Polley, and A. E. Hubbard (2007) · 2017
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…