We study identification and estimation of causal effects in settings with panel data. Traditionally researchers follow model-based identification strategies relying on assumptions governing the relation between the potential outcomes and the observed and unobserved confounders. We focus on a different, complementary approach to identification where assumptions are made about the connection between the treatment assignment and the unobserved confounders. Such strategies are common in cross-section settings but rarely used with panel data. We introduce different sets of assumptions that follow the two paths to identification and develop a doubly robust approach. We propose estimation methods that build on these identification strategies.
This paper studies a panel data setting where the goal is to estimate causal effects of an intervention by predicting the counterfactual values of outcomes for treated units, had they not received the treatment. Several approaches have been proposed for this problem, including regression methods, synthetic control methods and matrix completion methods. This paper considers an ensemble approach, and shows that it performs better than any of the individual methods in several economic datasets. Matrix completion methods are often given the most weight by the ensemble, but this clearly depends on …
The relationship between democracy and economic growth is of long-standing interest. We revisit the panel data analysis of this relationship by Acemoglu, Naidu, Restrepo and Robinson (forthcoming) using state of the art econometric methods. We argue that this and lots of other panel data settings in economics are in fact high-dimensional, resulting in principal estimators -- the fixed effects (FE) and Arellano-Bond (AB) estimators -- to be biased to the degree that invalidates statistical inference. We can however remove these biases by using simple analytical and sample-splitting methods, and…
Survey scientists increasingly face the problem of high-dimensionality in their research as digitization makes it much easier to construct high-dimensional (or "big") data sets through tools such as online surveys and mobile applications. Machine learning methods are able to handle such data, and they have been successfully applied to solve \emph{predictive} problems. However, in many situations, survey statisticians want to learn about \emph{causal} relationships to draw conclusions and be able to transfer the findings of one survey to another. Standard machine learning methods provide biased…
This article proposes doubly robust estimators for the average treatment effect on the treated (ATT) in difference-in-differences (DID) research designs. In contrast to alternative DID estimators, the proposed estimators are consistent if either (but not necessarily both) a propensity score or outcome regression working models are correctly specified. We also derive the semiparametric efficiency bound for the ATT in DID designs when either panel or repeated cross-section data are available, and show that our proposed estimators attain the semiparametric efficiency bound when the working models…
This paper develops the inferential theory for latent factor models estimated from large dimensional panel data with missing observations. We propose an easy-to-use all-purpose estimator for a latent factor model by applying principal component analysis to an adjusted covariance matrix estimated from partially observed panel data. We derive the asymptotic distribution for the estimated factors, loadings and the imputed values under an approximate factor model and general missing patterns. The key application is to estimate counterfactual outcomes in causal inference from panel data. The unobse…
We consider the problem of constructing bounds on the average treatment effect (ATE) when unmeasured confounders exist but have bounded influence. Specifically, we assume that omitted confounders could not change the odds of treatment for any unit by more than a fixed factor. We derive the sharp partial identification bounds implied by this assumption by leveraging distributionally robust optimization, and we propose estimators of these bounds with several novel robustness properties. The first is double sharpness: our estimators consistently estimate the sharp ATE bounds when one of two nuisa…
In this paper we study methods for estimating causal effects in settings with panel data, where some units are exposed to a treatment during some periods and the goal is estimating counterfactual (untreated) outcomes for the treated unit/period combinations. We propose a class of matrix completion estimators that uses the observed elements of the matrix of control outcomes corresponding to untreated unit/periods to impute the "missing" elements of the control outcome matrix, corresponding to treated units/periods. This leads to a matrix that well-approximates the original (incomplete) matrix, …
We develop a new approach for identifying and estimating average causal effects in panel data under a linear factor model with unmeasured confounders. Compared to other methods tackling factor models such as synthetic controls and matrix completion, our method does not require the number of time periods to grow infinitely. Instead, we draw inspiration from the two-way fixed effect model as a special case of the linear factor model, where a simple difference-in-differences transformation identifies the effect. We show that analogous, albeit more complex, transformations exist in the more genera…
Johns Hopkins University（约翰斯·霍普金斯大学）；University of Pennsylvania（宾夕法尼亚大学）
Robins et al. (2008) introduced a class of influence functions (IFs) which could be used to obtain doubly robust moment functions for the corresponding parameters. However, that class does not include the IF of parameters for which the nuisance functions are solutions to integral equations. Such parameters are particularly important in the field of causal inference, specifically in the recently proposed proximal causal inference framework of Tchetgen Tchetgen et al. (2020), which allows for estimating the causal effect in the presence of latent confounders. In this paper, we first extend the c…
Double Machine Learning for Static Panel Data with Instrumental Variables: New Method and Applications
Erasmus School of Economics, Erasmus University（伊拉斯谟斯大学经济研究生院）；Tinbergen Institute（丁伯根研究所）；Department of Business Analytics and Applied Economics, Queen Mary University of London（伦敦大学玛丽女王学院商业分析与应用经济学系）；Institute for Social and Economic Research, University of Essex（埃塞克斯大学社会与经济研究所）
本文提出静态面板模型中处理内生处理的双重机器学习估计器（panel IV DML），并引入弱识别诊断，通过三个迁移研究验证其在高维非线性影响下的稳健性与准确性。
Heterogeneous panel data models that allow the coefficients to vary across individuals and/or change over time have received increasingly more attention in statistics and econometrics. This paper proposes a two-dimensional heterogeneous panel regression model that incorporate a group structure of individual heterogeneous effects with cohort formation for their time-variations, which allows common coefficients between nonadjacent time points. A bi-integrative procedure that detects the information regarding group and cohort patterns simultaneously via a doubly penalized least square with concav…
Many econometrics textbooks imply that under mean independence of the regressors and the error term, the OLS parameters have a causal interpretation. We show that even when this assumption is satisfied, OLS might identify a pseudo-parameter that does not have a causal interpretation. Even assuming that the linear model is "structural" creates some ambiguity in what the regression error represents and whether the OLS estimand is causal. This issue applies equally to linear IV and panel data models. To give these estimands a causal interpretation, one needs to impose assumptions on a "causal" mo…
A central goal in social science is to evaluate the causal effect of a policy. One dominant approach is through panel data analysis in which the behaviors of multiple units are observed over time. The information across time and space motivates two general approaches: (i) horizontal regression (i.e., unconfoundedness), which exploits time series patterns, and (ii) vertical regression (e.g., synthetic controls), which exploits cross-sectional patterns. Conventional wisdom states that the two approaches are fundamentally different. We establish this position to be partly false for estimation but…
This paper proposes a new class of M-estimators that double weight for the twin problems of nonrandom treatment assignment and missing outcomes, both of which are common issues in the treatment effects literature. The proposed class is characterized by a `robustness' property, which makes it resilient to parametric misspecification in either a conditional model of interest (for example, mean or quantile function) or the two weighting functions. As leading applications, the paper discusses estimation of two specific causal parameters; average and quantile treatment effects (ATE, QTEs), which ca…
We introduce the Pricing Engine package to enable the use of Double ML estimation techniques in general panel data settings. Customization allows the user to specify first-stage models, first-stage featurization, second stage treatment selection and second stage causal-modeling. We also introduce a DynamicDML class that allows the user to generate dynamic treatment-aware forecasts at a range of leads and to understand how the forecasts will vary as a function of causally estimated treatment parameters. The Pricing Engine is built on Python 3.5 and can be run on an Azure ML Workbench environmen…
The doubly robust (DR) estimator, which consists of two nuisance parameters, the conditional mean outcome and the logging policy (the probability of choosing an action), is crucial in causal inference. This paper proposes a DR estimator for dependent samples obtained from adaptive experiments. To obtain an asymptotically normal semiparametric estimator from dependent samples with non-Donsker nuisance estimators, we propose adaptive-fitting as a variant of sample-splitting. We also report an empirical paradox that our proposed DR estimator tends to show better performances compared to other est…
