For a panel model considered by Abadie et al. (2010), the counterfactual outcomes constructed by Abadie et al., Hsiao et al. (2012), and Doudchenko and Imbens (2017) may all be confounded by uncontrolled heterogenous trends. Based on exact-matching on the trend predictors, I propose new methods of estimating the model-specific treatment effects, which are free from heterogenous trends. When applied to Abadie et al.'s (2010) model and data, the new estimators suggest considerably smaller effects of California's tobacco control program.
Nonseparable panel models are important in a variety of economic settings, including discrete choice. This paper gives identification and estimation results for nonseparable models under time homogeneity conditions that are like "time is randomly assigned" or "time is an instrument." Partial identification results for average and quantile effects are given for discrete regressors, under static or dynamic conditions, in fully nonparametric and in semiparametric models, with time effects. It is shown that the usual, linear, fixed-effects estimator is not a consistent estimator of the identified …
Learning Treatment Effects in Panels with General Intervention Patterns
Sloan School of Management（斯隆管理学院）；Tepper School of Business, Carnegie Mellon University（卡内基梅隆大学泰珀商学院）；Massachusetts Institute of Technology（麻省理工学院）；Department of Aeronautics and Astronautics, Massachusetts Institute of Technology（麻省理工学院航空航天系）
This article develops new closed-form variance expressions for power analyses for commonly used difference-in-differences (DID) and comparative interrupted time series (CITS) panel data estimators. The main contribution is to incorporate variation in treatment timing into the analysis. The power formulas also account for other key design features that arise in practice: autocorrelated errors, unequal measurement intervals, and clustering due to the unit of treatment assignment. We consider power formulas for both cross-sectional and longitudinal models and allow for covariates. An illustrative…
In order to evaluate the impact of a policy intervention on a group of units over time, it is important to correctly estimate the average treatment effect (ATE) measure. Due to lack of robustness of the existing procedures of estimating ATE from panel data, in this paper, we introduce a robust estimator of the ATE and the subsequent inference procedures using the popular approach of minimum density power divergence inference. Asymptotic properties of the proposed ATE estimator are derived and used to construct robust test statistics for testing parametric hypotheses related to the ATE. Besides…
This paper considers estimating functional-coefficient models in panel quantile regression with individual effects, allowing the cross-sectional and temporal dependence for large panel observations. A latent group structure is imposed on the heterogenous quantile regression models so that the number of nonparametric functional coefficients to be estimated can be reduced considerably. With the preliminary local linear quantile estimates of the subject-specific functional coefficients, a classic agglomerative clustering algorithm is used to estimate the unknown group structure and an easy-to-imp…
Practitioners in medicine, business, political science, and other fields are increasingly aware that decisions should be personalized to each patient, customer, or voter. A given treatment (e.g. a drug or advertisement) should be administered only to those who will respond most positively, and certainly not to those who will be harmed by it. Individual-level treatment effects can be estimated with tools adapted from machine learning, but different models can yield contradictory estimates. Unlike risk prediction models, however, treatment effect models cannot be easily evaluated against each ot…
The causal inference literature has increasingly recognized that explicitly targeting treatment effect heterogeneity can lead to improved scientific understanding and policy recommendations. Towards the same ends, studying the causal pathway connecting the treatment to the outcome can be also useful. This paper addresses these problems in the context of \emph{causal mediation analysis}. We introduce a varying coefficient model based on Bayesian additive regression trees to identify and regularize heterogeneous causal mediation effects; analogously with linear structural equation models, these …
This paper proposes a method for estimating the effect of a policy intervention on an outcome over time. We train recurrent neural networks (RNNs) on the history of control unit outcomes to learn a useful representation for predicting future outcomes. The learned representation of control units is then applied to the treated units for predicting counterfactual outcomes. RNNs are specifically structured to exploit temporal dependencies in panel data, and are able to learn negative and nonlinear interactions between control unit outcomes. We apply the method to the problem of estimating the long…
This paper provides estimation and inference methods for a conditional average treatment effects (CATE) characterized by a high-dimensional parameter in both homogeneous cross-sectional and unit-heterogeneous dynamic panel data settings. In our leading example, we model CATE by interacting the base treatment variable with explanatory variables. The first step of our procedure is orthogonalization, where we partial out the controls and unit effects from the outcome and the base treatment and take the cross-fitted residuals. This step uses a novel generic cross-fitting method we design for weakl…
Spillover of economic outcomes often arises over multiple networks, and distinguishing their separate roles is important in empirical research. For example, the direction of spillover between two groups (such as banks and industrial sectors linked in a bipartite graph) has important economic implications, and a researcher may want to learn which direction is supported in the data. For this, we need to have an empirical methodology that allows for both directions of spillover simultaneously. In this paper, we develop a dynamic linear panel model and asymptotic inference with large $n$ and small…
Marginal structural models are a popular tool for investigating the effects of time-varying treatments, but they require an assumption of no unobserved confounders between the treatment and outcome. With observational data, this assumption may be difficult to maintain, and in studies with panel data, many researchers use fixed effects models to purge the data of time-constant unmeasured confounding. Unfortunately, traditional linear fixed effects models are not suitable for estimating the effects of time-varying treatments, since they can only estimate lagged effects under implausible assumpti…
The synthetic control method (SCM) allows estimating the causal effect of an intervention in settings where panel data on a small number of treated and control units are available. We show that the existing SCM, as well as its extensions, can be easily modified to estimate how much of the ``total'' effect goes through observed causal channels. Our new mediation analysis synthetic control (MASC) method requires additional assumptions that are arguably mild in many settings. We illustrate the implementation of MASC in an empirical application estimating the direct and indirect effects of an anti…
Causal or unconfounded descriptive comparisons between multiple groups are common in observational studies. Motivated from a racial disparity study in health services research, we propose a unified propensity score weighting framework, the balancing weights, for estimating causal effects with multiple treatments. These weights incorporate the generalized propensity scores to balance the weighted covariate distribution of each treatment group, all weighted toward a common pre-specified target population. The class of balancing weights include several existing approaches such as the inverse prob…
Within the national innovation system literature, empirical analyses are severely lacking for developing economies. Particularly, the low- and middle-income countries (LMICs) eligible for the World Bank's International Development Association (IDA) support, are rarely part of any empirical discourse on growth, development, and innovation. One major issue hindering panel analyses in LMICs, and thus them being subject to any empirical discussion, is the lack of complete data availability. This work offers a new complete panel dataset with no missing values for LMICs eligible for IDA's support. I…
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…
This article develops a Bayesian approach for estimating panel quantile regression with binary outcomes in the presence of correlated random effects. We construct a working likelihood using an asymmetric Laplace (AL) error distribution and combine it with suitable prior distributions to obtain the complete joint posterior distribution. For posterior inference, we propose two Markov chain Monte Carlo (MCMC) algorithms but prefer the algorithm that exploits the blocking procedure to produce lower autocorrelation in the MCMC draws. We also explain how to use the MCMC draws to calculate the margin…
This paper considers generalized least squares (GLS) estimation for linear panel data models. By estimating the large error covariance matrix consistently, the proposed feasible GLS (FGLS) estimator is more efficient than the ordinary least squares (OLS) in the presence of heteroskedasticity, serial, and cross-sectional correlations. To take into account the serial correlations, we employ the banding method. To take into account the cross-sectional correlations, we suggest to use the thresholding method. We establish the limiting distribution of the proposed estimator. A Monte Carlo study is c…
We consider estimation and inference in panel data models with additive unobserved individual specific heterogeneity in a high dimensional setting. The setting allows the number of time varying regressors to be larger than the sample size. To make informative estimation and inference feasible, we require that the overall contribution of the time varying variables after eliminating the individual specific heterogeneity can be captured by a relatively small number of the available variables whose identities are unknown. This restriction allows the problem of estimation to proceed as a variable s…
