Open Access
2015 Some models and methods for the analysis of observational data
José A. Ferreira
Statist. Surv. 9: 106-208 (2015). DOI: 10.1214/15-SS110

Abstract

This article provides a concise and essentially self-contained exposition of some of the most important models and non-parametric methods for the analysis of observational data, and a substantial number of illustrations of their application. Although for the most part our presentation follows P. Rosenbaum’s book, “Observational Studies”, and naturally draws on related literature, it contains original elements and simplifies and generalizes some basic results. The illustrations, based on simulated data, show the methods at work in some detail, highlighting pitfalls and emphasizing certain subjective aspects of the statistical analyses.

Citation

Download Citation

José A. Ferreira. "Some models and methods for the analysis of observational data." Statist. Surv. 9 106 - 208, 2015. https://doi.org/10.1214/15-SS110

Information

Received: 1 June 2015; Published: 2015
First available in Project Euclid: 16 September 2015

zbMATH: 1327.62284
MathSciNet: MR3396384
Digital Object Identifier: 10.1214/15-SS110

Subjects:
Primary: 62G05 , 62G10 , 62G15 , 62G99
Secondary: 62P10 , 62P25

Keywords: confounding , estimation , observational data , propensity scores , stratification , testing , treatment effect

Rights: Copyright © 2015 The author, under a Creative Commons Attribution License

Vol.9 • 2015
Back to Top