PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
June 1, 1992The Annals of StatisticsOpen Access

Affinely Invariant Matching Methods with Ellipsoidal Distributions

View Full Paper
Ask AI
Bookmark
Share

Authors

DRDonald B. RubinNTNeal Thomas

Discussion

Loading...

Member takes

Overview

Randomized trial explores bias reduction methods in observational studies, suggesting new theoretical frameworks.

Key Points

  • This research aims to establish a theoretical framework for evaluating affinely invariant matching methods in observational studies.
  • Developed a general theoretical framework for affinely invariant matching methods.
  • Examined performance concerning ellipsoidal distributions and bias reduction.
  • Analyzed conditionally affinely invariant methods for covariates with ellipsoidal distributions.
  • Identified performance characteristics of matching methods in a subspace aligned with the best linear discriminant.
  • Demonstrated how these methods can decompose matching effects into relevant subspaces.
  • Provided theoretical foundations for matched sampling utilizing estimated propensity scores.

Cite This Study

Rubin et al. (1992) studied this question.

synapsesocial.com/papers/6a12be4d257f24f1de9e36dchttps://doi.org/10.1214/aos/1176348671
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Multivariate Matching Methods That are Equal Percent Bias Reducing, II: Maximums on Bias Reduction for Fixed Sample Sizes1976 · 125 citations
  2. 2Bias mitigation in matched observational studies with continuous treatments: calipered non-bipartite matching and bias-corrected estimation and inference2026
  3. 3Variance Estimation in Matched Difference‐in‐Differences Designs2026
  4. 4Effects of Matching on Evaluation of Accuracy, Fairness, and Fairness Impossibility in AI-ML Systems2024 · 1 citations
  5. 5Augmented match weighted estimators: new methods for estimating average treatment effects under extreme propensity scores2026