Matching is an R package which provides functions for multivariate and propensity score matching and for finding optimal covariate balance based on a genetic search algorithm. A variety of univariate ...and multivariate metrics to determine if balance actually has been obtained are provided. The underlying matching algorithm is written in C++, makes extensive use of system BLAS and scales efficiently with dataset size. The genetic algorithm which finds optimal balance is parallelized and can make use of multiple CPUs or a cluster of computers. A large number of options are provided which control exactly how the matching is conducted and how balance is evaluated.
This paper presents genetic matching, a method of multivariate matching that uses an evolutionary search algorithm to determine the weight each covariate is given. Both propensity score matching and ...matching based on Mahalanobis distance are limiting cases of this method. The algorithm makes transparent certain issues that all matching methods must confront. We present simulation studies that show that the algorithm improves covariate balance and that it may reduce bias if the selection on observables assumption holds. We then present a reanalysis of a number of data sets in the LaLonde (1986) controversy.
There is growing interest in estimating and analyzing heterogeneous treatment effects in experimental and observational studies. We describe a number of metaalgorithms that can take advantage of any ...supervised learning or regression method in machine learning and statistics to estimate the conditional average treatment effect (CATE) function. Metaalgorithms build on base algorithms—such as random forests (RFs), Bayesian additive regression trees (BARTs), or neural networks—to estimate the CATE, a function that the base algorithms are not designed to estimate directly. We introduce a metaalgorithm, the X-learner, that is provably efficient when the number of units in one treatment group is much larger than in the other and can exploit structural properties of the CATE function. For example, if the CATE function is linear and the response functions in treatment and control are Lipschitz-continuous, the X-learner can still achieve the parametric rate under regularity conditions. We then introduce versions of the X-learner that use RF and BART as base learners. In extensive simulation studies, the X-learner performs favorably, although none of the metalearners is uniformly the best. In two persuasion field experiments from political science, we demonstrate how our X-learner can be used to target treatment regimes and to shed light on underlying mechanisms. A software package is provided that implements our methods.
genoud is an R function that combines evolutionary algorithm methods with a derivative-based (quasi-Newton) method to solve difficult optimization problems. genoud may also be used for optimization ...problems for which derivatives do not exist. genoud solves problems that are nonlinear or perhaps even discontinuous in the parameters of the function to be optimized. When the function to be optimized (for example, a log-likelihood) is nonlinear in the model's parameters, the function will generally not be globally concave and may have irregularities such as saddlepoints or discontinuities. Optimization methods that rely on derivatives of the objective function may be unable to find any optimum at all. Multiple local optima may exist, so that there is no guarantee that a derivative-based method will converge to the global optimum. On the other hand, algorithms that do not use derivative information (such as pure genetic algorithms) are for many problems needlessly poor at local hill climbing. Most statistical problems are regular in a neighborhood of the solution. Therefore, for some portion of the search space, derivative information is useful. The function supports parallel processing on multiple CPUs on a single machine or a cluster of computers.
In recent years, there has been a burst of innovative work on methods for estimating causal effects using observational data. Much of this work has extended and brought a renewed focus on old ...approaches such as matching, which is the focus of this review. The new developments highlight an old tension in the social sciences: a focus on research design versus a focus on quantitative models. This realization, along with the renewed interest in field experiments, has marked the return of foundational questions as opposed to a fascination with the latest estimator. I use studies of get-out-the-vote interventions to exemplify this development. Without an experiment, natural experiment, a discontinuity, or some other strong design, no amount of econometric or statistical modeling can make the move from correlation to causation persuasive.
We provide a principled way for investigators to analyze randomized experiments when the number of covariates is large. Investigators often use linear multivariate regression to analyze randomized ...experiments instead of simply reporting the difference of means between treatment and control groups. Their aim is to reduce the variance of the estimated treatment effect by adjusting for covariates. If there are a large number of covariates relative to the number of observations, regression may perform poorly because of overfitting. In such cases, the least absolute shrinkage and selection operator (Lasso) may be helpful. We study the resulting Lasso-based treatment effect estimator under the Neyman–Rubin model of randomized experiments. We present theoretical conditions that guarantee that the estimator is more efficient than the simple difference-of-means estimator, and we provide a conservative estimator of the asymptotic variance, which can yield tighter confidence intervals than the difference-of-means estimator. Simulation and data examples show that Lasso-based adjustment can be advantageous even when the number of covariates is less than the number of observations. Specifically, a variant using Lasso for selection and ordinary least squares (OLS) for estimation performs particularly well, and it chooses a smoothing parameter based on combined performance of Lasso and OLS.
Following David Lee's pioneering work, numerous scholars have applied the regression discontinuity (RD) design to popular elections. Contrary to the assumptions of RD, however, we show that bare ...winners and bare losers in U.S. House elections (1942–2008) differ markedly on pretreatment covariates. Bare winners possess large ex ante financial, experience, and incumbency advantages over their opponents and are usually the candidates predicted to win by Congressional Quarterly's pre-election ratings. Covariate imbalance actually worsens in the closest House elections. National partisan tides help explain these patterns. Previous works have missed this imbalance because they rely excessively on model-based extrapolation. We present evidence suggesting that sorting in close House elections is due mainly to activities on or before Election Day rather than postelection recounts or other manipulation. The sorting is so strong that it is impossible to achieve covariate balance between matched treated and control observations, making covariate adjustment a dubious enterprise. Although RD is problematic for postwar House elections, this example does highlight the design's advantages over alternatives: RD's assumptions are clear and weaker than model-based alternatives, and their implications are empirically testable.
Natural experiments help to overcome some of the obstacles researchers face when making causal inferences in the social sciences. However, even when natural interventions are randomly assigned, some ...of the treatment–control comparisons made available by natural experiments may not be valid. We offer a framework for clarifying the issues involved, which are subtle and often overlooked. We illustrate our framework by examining four different natural experiments used in the literature. In each case, random assignment of the intervention is not sufficient to provide an unbiased estimate of the causal effect. Additional assumptions are required that are problematic. For some examples, we propose alternative research designs that avoid these conceptual difficulties.
We derive new variance formulas for inference on a general class of estimands of causal average treatment effects in a randomized control trial. We generalize the seminal work of Robins and show that ...when the researcher's objective is inference on sample average treatment effect of the treated (SATT), a consistent variance estimator exists. Although this estimand is equal to the sample average treatment effect (SATE) in expectation, potentially large differences in both accuracy and coverage can occur by the change of estimand, even asymptotically. Inference on SATE, even using a conservative confidence interval, provides incorrect coverage of SATT. We demonstrate the applicability of the new theoretical results using an empirical application with hundreds of online experiments with an average sample size of approximately 100 million observations per experiment. An R package, estCI, that implements all the proposed estimation procedures is available.
Supplementary materials
for this article are available online.
Randomized controlled trials (RCTs) can provide unbiased estimates of sample average treatment effects. However, a common concern is that RCTs may fail to provide unbiased estimates of population ...average treatment effects. We derive the assumptions that are required to identify population average treatment effects from RCTs. We provide placebo tests, which formally follow from the identifying assumptions and can assess whether they hold. We offer new research designs for estimating population effects that use non-randomized studies to adjust the RCT data. This approach is considered in a cost-effectiveness analysis of a clinical intervention: pulmonary artery catheterization.