Statistical Models and Causal Inference: Weighting Regressions by Propensity Scores

By R. Berk. 2009

217 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/d6da1e4287da2511c141061d3da35088c4647de3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Comparison of Approaches to Weight Truncation for Marginal Structural Cox Models (2013, background): “It would be interesting to investigate, in future research, whether the approaches considered in our study would improve the performance (bias and/or variance) of the IPTW estimators in the situation where estimates using untruncated weights are themselves biased (Freedman and Berk, 2008), and in si”
What Do We Know About Gangs and Gang Members and Where Do We Go From Here? (2013, methodology): “While these methods are by no means a panacea with regard to causal inference (see e.g. Bjerk, 2009; Freedman & Berk, 2008 for cautionary notes), when utilized correctly they appear to produce more reliable estimates of the true gang effect.”
Covariate balancing propensity score (2014, methodology): “The bias and RMSE are †The bias and RMSE are computed for the Horvitz–Thompson HT, the inverse propensity score weighting IPW, the inverse propensity score weighted least squares WLS and the doubly robust least squares DR estimators.”
Residual Unfairness in Fair Machine Learning from Prejudiced Data (2018, methodology): “Sampling adjustment and re-weighting is commonly used in the social sciences, medicine, and epidemiology for ensuring the validity of population-level inference where there is population mismatch between studies and the population of interest (Thompson, 2012; Freedman & Berk, 2008).”
Genetic Matching for Estimating Causal Effects: A General Multivariate Matching Method for Achieving Balance in Observational Studies (2006, background): “However, there are open questions about the fragility of such weighting estimators in finite samples when the estimated probabilities of treatment assignment are close to 0 or 1 (Freedman & Berk, 2008; Kang & Schafer, 2007; Porter et al., 2011).”