We consider the problem of A-B testing when the impact of the treatment is marred by a large number of covariates. Randomization can be highly inefficient in such settings, and thus we consider the problem of optimally allocating test subjects to either treatment with a view to maximizing the precision of our estimate of the treatment effect. Our main contribution is a tractable algorithm for this problem in the online setting, where subjects arrive, and must be assigned, sequentially, with covariates drawn from an elliptical distribution with finite second moment. We further characterize the gain in precision afforded by optimized allocations relative to randomized allocations, and show that this gain grows large as the number of covariates grows. Our dynamic optimization framework admits several generalizations that incorporate important operational constraints such as the consideration of selection bias, budgets on allocations, and endogenous stopping times. In a set of numerical experiments, we demonstrate that our method simultaneously offers better statistical efficiency and less selection bias than state-of-the-art competing biased coin designs.
Minimizing Interference and Selection Bias in Network Experiment Design
Current approaches to A/B testing in networks focus on limiting interference, the concern that treatment effects can ”spill over” from treatment nodes to control nodes and lead to biased causal effect estimation. Prominent methods for network experiment design rely on two-stage randomization, in which sparsely-connected clusters are identified and cluster randomization dictates the node assignment to treatment and control. Here, we show that cluster randomization does not ensure sufficient node randomization and it can lead to selection bias in which treatment and control nodes represent different populations of users. To address this problem, we propose a principled framework for network experiment design which jointly minimizes interference and selection bias. We introduce the concepts of edge spillover probability and cluster matching and demonstrate their importance for designing network A/B testing. Our experiments on a number of real-world datasets show that our proposed framework leads to significantly lower error in causal effect estimation than existing solutions.
- Award ID(s):
- 1801644
- Publication Date:
- NSF-PAR ID:
- 10230876
- Journal Name:
- 14th International AAAI Conference on Web and Social Media
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
For large observational studies lacking a control group (unlike randomized controlled trials, RCT), propensity scores (PS) are often the method of choice to account for pre-treatment confounding in baseline characteristics, and thereby avoid substantial bias in treatment estimation. A vast majority of PS techniques focus on average treatment effect estimation, without any clear consensus on how to account for confounders, especially in a multiple treatment setting. Furthermore, for time-to event outcomes, the analytical framework is further complicated in presence of high censoring rates (sometimes, due to non-susceptibility of study units to a disease), imbalance between treatment groups, and clustered nature of the data (where, survival outcomes appear in groups). Motivated by a right-censored kidney transplantation dataset derived from the United Network of Organ Sharing (UNOS), we investigate and compare two recent promising PS procedures, (a) the generalized boosted model (GBM), and (b) the covariate-balancing propensity score (CBPS), in an attempt to decouple the causal effects of treatments (here, study subgroups, such as hepatitis C virus (HCV) positive/negative donors, and positive/negative recipients) on time to death of kidney recipients due to kidney failure, post transplantation. For estimation, we employ a 2-step procedure which addresses various complexities observed in the UNOS databasemore »
-
Abstract Standard estimators of the global average treatment effect can be biased in the presence of interference. This paper proposes regression adjustment estimators for removing bias due to interference in Bernoulli randomized experiments. We use a fitted model to predict the counterfactual outcomes of global control and global treatment. Our work differs from standard regression adjustments in that the adjustment variables are constructed from functions of the treatment assignment vector, and that we allow the researcher to use a collection of any functions correlated with the response, turning the problem of detecting interference into a feature engineering problem. We characterize the distribution of the proposed estimator in a linear model setting and connect the results to the standard theory of regression adjustments under SUTVA. We then propose an estimator that allows for flexible machine learning estimators to be used for fitting a nonlinear interference functional form. We propose conducting statistical inference via bootstrap and resampling methods, which allow us to sidestep the complicated dependences implied by interference and instead rely on empirical covariance structures. Such variance estimation relies on an exogeneity assumption akin to the standard unconfoundedness assumption invoked in observational studies. In simulation experiments, our methods are better atmore »
-
In this work, we study the optimal design of two-armed clinical trials to maximize the accuracy of parameter estimation in a statistical model, where the interaction between patient covariates and treatment are explicitly incorporated to enable precision medication decisions. Such a modeling extension leads to significant complexities for the produced optimization problems because they include optimization over design and covariates concurrently. We take a min-max optimization model and minimize (over design) the maximum (over population) variance of the estimated interaction effect between treatment and patient covariates. This results in a min-max bilevel mixed integer nonlinear programming problem, which is notably challenging to solve. To address this challenge, we introduce a surrogate optimization model by approximating the objective function, for which we propose two solution approaches. The first approach provides an exact solution based on reformulation and decomposition techniques. In the second approach, we provide a lower bound for the inner optimization problem and solve the outer optimization problem over the lower bound. We test our proposed algorithms with synthetic and real-world data sets and compare them with standard (re)randomization methods. Our numerical analysis suggests that the proposed approaches provide higher-quality solutions in terms of the variance of estimators and probabilitymore »
-
Recent work on legislative politics has documented complex patterns of interaction and collaboration through the lens of network analysis. In a largely separate vein of research, the field experiment—with many applications in state legislatures—has emerged as an important approach in establishing causal identification in the study of legislative politics. The stable unit treatment value assumption (SUTVA)—the assumption that a unit’s outcome is unaffected by other units’ treatment statuses—is required in conventional approaches to causal inference with experiments. When SUTVA is violated via networked social interaction, treatment effects spread to control units through the network structure. We review recently developed methods that can be used to account for interference in the analysis of data from field experiments on state legislatures. The methods we review require the researcher to specify a spillover model, according to which legislators influence each other, and specify the network through which spillover occurs. We discuss these and other specification steps in detail. We find mixed evidence for spillover effects in data from two previously published field experiments. Our replication analyses illustrate how researchers can use recently developed methods to test for interference effects, and support the case for considering interference effects in experiments on state legislatures.
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Publisher's Version of Record
- The DOI is not currently available
