Covariate Balancing Propensity Score Research Articles

Robust Estimation of Inverse Probability Weights for Marginal Structural Models

Marginal structural models (MSMs) are becoming increasingly popular as a tool for causal inference from longitudinal data. Unlike standard regression models, MSMs can adjust for time-dependent observed confounders while avoiding the bias due to the direct adjustment for covariates affected by the treatment. Despite their theoretical appeal, a main practical difficulty of MSMs is the required estimation of inverse probability weights. Previous studies have found that MSMs can be highly sensitive to misspecification of treatment assignment model even when the number of time periods is moderate. To address this problem, we generalize the covariate balancing propensity score (CBPS) methodology of Imai and Ratkovic to longitudinal analysis settings. The CBPS estimates the inverse probability weights such that the resulting covariate balance is improved. Unlike the standard approach, the proposed methodology incorporates all covariate balancing conditions across multiple time periods. Since the number of these conditions grows exponentially as the number of time period increases, we also propose a low-rank approximation to ease the computational burden. Our simulation and empirical studies suggest that the CBPS significantly improves the empirical performance of MSMs by making the treatment assignment model more robust to misspecification. Open-source software is available for implementing the proposed methods.

Impact of remittances on poverty at origin: a study on rural households in India using Covariate Balancing Propensity Score Matching

In India, migration research has chiefly concentrated upon determining the causes rather than the effects of migration. A review of the extant literature reveals a near absence of studies exploring the linkages between remittances and poverty in India. This paper attempts to partly fill that void by assessing the effect of remittances on rural development. The National Sample Survey Office (NSSO), which carried out a nationwide survey on migration in 2007–2008, reported that out-migrants from rural India have a much higher propensity to remit compared to their urban counterparts. As rural poverty in India continues to pose the most fundamental development challenge, this paper explores on the basis of unit level data obtained from the NSSO survey whether transfer income in the form of remittances help to alleviate/de-intensify poverty among recipient rural households. Using the nascent technique of covariate balancing propensity score, rural households receiving remittances are matched with households that have similar background characteristics but do not receive remittances, and the impact of remittance income on their poverty status is subsequently ascertained. After correcting for self-selection, the study finds that remittance income from both internal as well as international sources do serve to lower the incidence of poverty among rural households although, expectedly, international remittances seem to have a stronger poverty alleviating effect.

Nominating Commissions, Judicial Retention, and Forward-Looking Behavior on State Supreme Courts: An Empirical Examination of Selection and Retention Methods

AbstractHigh-profile advocates are pushing states to move away from judicial elections and toward a “merit” method because it purportedly produces the best quality judges. Quality, however, is difficult to measure empirically. Rather than attempt to measure quality, we examine whether certain types of state supreme courts are more forward-looking than others. States are likely to desire forward-looking behavior among judges because it can protect judicial legitimacy, help states to control policy, and could be more efficient than myopic behavior. Using a recent innovation in matching called covariate-balancing propensity scores, we find that the U.S. Supreme Court is equally likely to review and reverse decisions by judges regardless of their selection or retention methods. These results suggest that state supreme court justices, no matter their paths of getting to (and staying on) their courts, are roughly equal in terms of forward-looking behavior.

Covariate Balancing Propensity Score

SummaryThe propensity score plays a central role in a variety of causal inference settings. In particular, matching and weighting methods based on the estimated propensity score have become increasingly common in the analysis of observational data. Despite their popularity and theoretical appeal, the main practical difficulty of these methods is that the propensity score must be estimated. Researchers have found that slight misspecification of the propensity score model can result in substantial bias of estimated treatment effects. We introduce covariate balancing propensity score (CBPS) methodology, which models treatment assignment while optimizing the covariate balance. The CBPS exploits the dual characteristics of the propensity score as a covariate balancing score and the conditional probability of treatment assignment. The estimation of the CBPS is done within the generalized method-of-moments or empirical likelihood framework. We find that the CBPS dramatically improves the poor empirical performance of propensity score matching and weighting methods reported in the literature. We also show that the CBPS can be extended to other important settings, including the estimation of the generalized propensity score for non-binary treatments and the generalization of experimental estimates to a target population. Open source software is available for implementing the methods proposed.