Literature DB >> 31982245

Visualization tool of variable selection in bias-variance tradeoff for inverse probability weights.

Ya-Hui Yu1, Kristian B Filion2, Lisa M Bodnar3, Maria M Brooks4, Robert W Platt5, Katherine P Himes6, Ashley I Naimi7.   

Abstract

PURPOSE: Inversed probability weighted (IPW) estimators are commonly used to adjust for time-fixed or time-varying confounders. However, in high-dimensional settings, including all identified confounders may result in unstable weights leading to higher variance. We aimed to develop a visualization tool demonstrating the impact of each confounder on the bias and variance of IPW estimates, as well as the propensity score overlap.
METHODS: A SAS macro was developed for this visualization tool and we demonstrate how this tool can be used to identify potentially problematic confounders of the association of statin use after myocardial infarction on one-year mortality in a plasmode simulation study using a cohort of 39,792 patients from the UK (1998-2012).
RESULTS: Through the tool's output, we can identify problematic confounders (two instrumental variables) and important confounders by comparing the estimated psuedo MSE with that from the fully adjusted model and propensity score overlap plot.
CONCLUSION: Our results suggest that the analytic impact of all confounders should be considered carefully when fitting IPW estimators.
Copyright © 2019. Published by Elsevier Inc.

Entities:  

Keywords:  Bias-variance tradeoff; High-dimensional covariates; Inverse probability weighting; Visualization

Mesh:

Year:  2019        PMID: 31982245      PMCID: PMC7864095          DOI: 10.1016/j.annepidem.2019.12.006

Source DB:  PubMed          Journal:  Ann Epidemiol        ISSN: 1047-2797            Impact factor:   3.797


  13 in total

1.  Marginal structural models and causal inference in epidemiology.

Authors:  J M Robins; M A Hernán; B Brumback
Journal:  Epidemiology       Date:  2000-09       Impact factor: 4.822

2.  Collaborative double robust targeted maximum likelihood estimation.

Authors:  Mark J van der Laan; Susan Gruber
Journal:  Int J Biostat       Date:  2010-05-17       Impact factor: 0.968

3.  The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trials.

Authors:  Donald B Rubin
Journal:  Stat Med       Date:  2007-01-15       Impact factor: 2.373

4.  Invited commentary: positivity in practice.

Authors:  Daniel Westreich; Stephen R Cole
Journal:  Am J Epidemiol       Date:  2010-02-05       Impact factor: 4.897

5.  An application of collaborative targeted maximum likelihood estimation in causal inference and genomics.

Authors:  Susan Gruber; Mark J van der Laan
Journal:  Int J Biostat       Date:  2010-05-17       Impact factor: 0.968

Review 6.  Statistical foundations for model-based adjustments.

Authors:  Sander Greenland; Neil Pearce
Journal:  Annu Rev Public Health       Date:  2015-03-18       Impact factor: 21.981

7.  Plasmode simulation for the evaluation of pharmacoepidemiologic methods in complex healthcare databases.

Authors:  Jessica M Franklin; Sebastian Schneeweiss; Jennifer M Polinski; Jeremy A Rassen
Journal:  Comput Stat Data Anal       Date:  2014-04       Impact factor: 1.681

8.  Constructing inverse probability weights for marginal structural models.

Authors:  Stephen R Cole; Miguel A Hernán
Journal:  Am J Epidemiol       Date:  2008-08-05       Impact factor: 4.897

Review 9.  Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies.

Authors:  Peter C Austin; Elizabeth A Stuart
Journal:  Stat Med       Date:  2015-08-03       Impact factor: 2.373

Review 10.  Outcome modelling strategies in epidemiology: traditional methods and basic alternatives.

Authors:  Sander Greenland; Rhian Daniel; Neil Pearce
Journal:  Int J Epidemiol       Date:  2016-04-20       Impact factor: 7.196

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.