We propose a new, flexible model for inference of the effect of a binary treatment on a continuous outcome observed over subsequent time periods. The model allows to seperate association due to endogeneity of treatment selection from additional longitudinal association of the outcomes and hence unbiased estimation of dynamic treatment effects.
We investigate the performance of the proposed method on simulated data and employ it to reanalyse data on the longitudinal effects of a long maternity leave
on mothers' earnings after their return to the labour market.
This study proposes a new Bayesian approach to infer binary treatment effects. The approach treats counterfactual untreated outcomes as missing observations and infers them by completing a matrix composed of realized and potential untreated outcomes using a data augmentation technique. We also develop a tailored prior that helps in the identification of parameters and induces the matrix of untreated outcomes to be approximately low rank. Posterior draws are simulated using a Markov Chain Monte Carlo sampler. While the proposed approach is similar to synthetic control methods and other related …
We are interested in the distribution of treatment effects for an experiment where units are randomized to a treatment but outcomes are measured for pairs of units. For example, we might measure risk sharing links between households enrolled in a microfinance program, employment relationships between workers and firms exposed to a trade shock, or bids from bidders to items assigned to an auction format. Such a double randomized experimental design may be appropriate when there are social interactions, market externalities, or other spillovers across units assigned to the same treatment. Or it …
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.
In panel experiments, we randomly assign units to different interventions, measuring their outcomes, and repeating the procedure in several periods. Using the potential outcomes framework, we define finite population dynamic causal effects that capture the relative effectiveness of alternative treatment paths. For a rich class of dynamic causal effects, we provide a nonparametric estimator that is unbiased over the randomization distribution and derive its finite population limiting distribution as either the sample size or the duration of the experiment increases. We develop two methods for i…
Researchers are often challenged with assessing the impact of an intervention on an outcome of interest in situations where the intervention is non-randomised, the intervention is only applied to one or few units, the intervention is binary, and outcome measurements are available at multiple time points. In this paper, we review existing methods for causal inference in these situations. We detail the assumptions underlying each method, emphasize connections between the different approaches and provide guidelines regarding their practical implementation. Several open problems are identified thu…
The problem we consider considers estimating a multivariate longitudinal panel data model whose outcomes can be a combination of discrete and continuous variables. This problem is challenging because the likelihood is usually analytically intractable. Our article makes both a methodological contribution and also a substantive contribution to the application. The methodological contribution is to introduce into the panel data literature a particle Metropolis within Gibbs method to carry out Bayesian inference, using a Hamiltonian Monte Carlo (Neal 2011} proposal for sampling the vector of unkno…
This paper constructs a doubly robust estimator for continuous dose-response estimation. An outcome regression model is augmented with a set of inverse generalized propensity score covariates to correct for potential misspecification bias. From the augmented model we can obtain consistent estimates of mean average potential outcomes for distinct strata of the treatment. A polynomial regression is then fitted to these point estimates to derive a Taylor approximation to the continuous dose-response function. The bootstrap is used for variance estimation. Analytical results and simulations show t…
In this paper, we adopt results in nonlinear time series analysis for causal inference in dynamical settings.~Our motivation is policy analysis with panel data, particularly through the use of "synthetic control" methods. These methods regress pre-intervention outcomes of the treated unit to outcomes from a pool of control units, and then use the fitted regression model to estimate causal effects post-intervention. In this setting, we propose to screen out control units that have a weak dynamical relationship to the treated unit. In simulations, we show that this method can mitigate bias from …
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…
A Causal Inference Approach for Evaluating Diagnostic Tests and AI-Enabled Medical Devices: From Effect Modification to Information-Augmented Decision-Making
Division of Biostatistics, University of California, Berkeley（加州大学伯克利分校生物统计系）
Department of Economics, University of North Carolina – Chapel Hill（北卡罗来纳大学教堂山分校经济学系）；European Commission, Joint Research Centre (JRC)（欧盟委员会联合研究中心）；Department of Finance, Copenhagen Business School（哥本哈根商学院金融系）
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…
Gilead Sciences, Inc.（吉利德科学公司）；Fred Hutchinson Cancer Center（弗雷德·哈钦森癌症中心）；University of Washington（华盛顿大学）
Heterogeneous treatment effects (HTE) based on patients' genetic or clinical factors are of significant interest to precision medicine. Simultaneously modeling HTE and corresponding main effects for randomized clinical trials with high-dimensional predictive markers is challenging. Motivated by the modified covariates approach, we propose a two-stage statistical learning procedure for estimating HTE with optimal efficiency augmentation, generalizing to arbitrary interaction model and exploiting powerful extreme gradient boosting trees (XGBoost). Target estimands for HTE are defined in the scal…
Clustering-Based Outcome Models for Clinical Studies: A Scoping Review
Medical University of Vienna（维也纳医科大学）；Hoffmann-La Roche AG（罗氏制药）；University of Warwick（华威大学）；Radboud University Medical Center（拉德堡德大学医学中心）；Bayer BCC（拜耳 BCC）；Inserm, Université Paris Cité, Inria, HeKA（法国国家健康与医学研究院、巴黎西岱大学、法国国家数字科学研究所、HeKA）
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…
Outcome-Aware Spectral Feature Learning for Instrumental Variable Regression
Gatsby Computational Neuroscience Unit, University College London（Gatsby计算神经科学单位，伦敦大学学院）；Faculty of Science, University of Novi Sad（科学学院，诺维萨德大学）；AI Centre, University College London（人工智能中心，伦敦大学学院）；DIBRIS, University of Genoa（DIBRIS，热那亚大学）
Difference-in-differences (DID) is a widely used approach for drawing causal inference from observational panel data. Two common estimation strategies for DID are outcome regression and propensity score weighting. In this paper, motivated by a real application in traffic safety research, we propose a new double-robust DID estimator that hybridizes regression and propensity score weighting. We particularly focus on the case of discrete outcomes. We show that the proposed double-robust estimator possesses the desirable large-sample robustness property. We conduct a simulation study to examine it…
This paper explores the identification and estimation of nonseparable panel data models. We show that the structural function is nonparametrically identified when it is strictly increasing in a scalar unobservable variable, the conditional distributions of unobservable variables do not change over time, and the joint support of explanatory variables satisfies some weak assumptions. To identify the target parameters, existing studies assume that the structural function does not change over time, and that there are "stayers", namely individuals with the same regressor values in two time periods.…
Precision medicine is of considerable interest in clinical, academic and regulatory parties. The key to precision medicine is the optimal treatment regime. Recently, Zhou et al. (2017) developed residual weighted learning (RWL) to construct the optimal regime that directly optimize the clinical outcome. However, this method involves computationally intensive non-convex optimization, which cannot guarantee a global solution. Furthermore, this method does not possess fully semiparametrical efficiency. In this article, we propose augmented outcome-weighted learning (AOL). The method is built on a…
In large epidemiologic studies, it is typical for an inexpensive, non-invasive procedure to be used to record disease status during regular follow-up visits, with less frequent assessment by a gold standard test. Inexpensive outcome measures like self-reported disease status are practical to obtain, but can be error-prone. Association analysis reliant on error-prone outcomes may lead to biased results; however, restricting analyses to only data from the less frequently observed error-free outcome could be inefficient. We have developed an augmented likelihood that incorporates data from both e…
A comparative study of augmented inverse propensity weighted estimators using outcome-oriented covariate selection via penalization with outcome-adaptive lasso
The stochastic frontier model with heterogeneous technical efficiency explained by exoge-nous variables is augmented with a spatial-temporal component, a generalization relaxing the panel independence assumption in a panel data. The estimation procedure takes advantage of additivity in the model, computational advantages over maximum likelihood estimation of parameters is exhibited. The spatial-temporal component can improve estimates of technical efficiency in a production frontier that is usually biased downwards. We present a test to veri-fy model assumptions that facilitates estimation of …
This paper provides a method to construct simultaneous confidence bands for quantile functions and quantile effects in nonlinear network and panel models with unobserved two-way effects, strictly exogenous covariates, and possibly discrete outcome variables. The method is based upon projection of simultaneous confidence bands for distribution functions constructed from fixed effects distribution regression estimators. These fixed effects estimators are debiased to deal with the incidental parameter problem. Under asymptotic sequences where both dimensions of the data set grow at the same rate,…
