• Bayesian Causality

    Type Journal Article
    Author Pierre Baldi
    Author Babak Shahbaba
    URL https://doi.org/10.1080/00031305.2019.1647876
    Volume 0
    Issue 0
    Pages 1-9
    Publication The American Statistician
    ISSN 0003-1305
    Date August 12, 2019
    DOI 10.1080/00031305.2019.1647876
    Accessed 8/30/2019, 11:14:58 AM
    Library Catalog Taylor and Francis+NEJM
    Abstract Although no universally accepted definition of causality exists, in practice one is often faced with the question of statistically assessing causal relationships in different settings. We present a uniform general approach to causality problems derived from the axiomatic foundations of the Bayesian statistical framework. In this approach, causality statements are viewed as hypotheses, or models, about the world and the fundamental object to be computed is the posterior distribution of the causal hypotheses, given the data and the background knowledge. Computation of the posterior, illustrated here in simple examples, may involve complex probabilistic modeling but this is no different than in any other Bayesian modeling situation. The main advantage of the approach is its connection to the axiomatic foundations of the Bayesian framework, and the general uniformity with which it can be applied to a variety of causality settings, ranging from specific to general cases, or from causes of effects to effects of causes.
    Date Added 8/30/2019, 11:14:58 AM
    Modified 8/30/2019, 11:15:43 AM

    Tags:

    • bayes
    • causal-inference
    • causal-model
    • causality
  • Bayesian sensitivity analysis for causal effects from 2 tables in the presence of unmeasured confounding with application to presidential campaign visits

    Type Journal Article
    Author Luke Keele
    Author Kevin M. Quinn
    URL https://doi.org/10.1214/17-AOAS1048
    Volume 11
    Issue 4
    Pages 1974-1997
    Publication Ann App Stat
    Date 2017-12
    Extra Citation Key: kee17bay tex.citeulike-article-id= 14546623 tex.citeulike-linkout-0= http://dx.doi.org/10.1214/17-AOAS1048 tex.citeulike-linkout-1= https://doi.org/10.1214/17-AOAS1048 tex.posted-at= 2018-03-09 20:04:17 tex.priority= 2
    DOI 10.1214/17-AOAS1048
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • bayesian-inference
    • bayes
    • confounding
    • sensitivity-analysis
    • causal-inference
  • The Central Role of Bayes’ Theorem for Joint Estimation of Causal Effects and Propensity Scores

    Type Journal Article
    Author Corwin Matthew Zigler
    URL https://doi.org/10.1080/00031305.2015.1111260
    Volume 70
    Issue 1
    Pages 47-54
    Publication The American Statistician
    ISSN 0003-1305
    Date January 2, 2016
    Extra PMID: 27482121
    DOI 10.1080/00031305.2015.1111260
    Accessed 11/25/2019, 7:36:13 AM
    Library Catalog Taylor and Francis+NEJM
    Abstract Although propensity scores have been central to the estimation of causal effects for over 30 years, only recently has the statistical literature begun to consider in detail methods for Bayesian estimation of propensity scores and causal effects. Underlying this recent body of literature on Bayesian propensity score estimation is an implicit discordance between the goal of the propensity score and the use of Bayes’ theorem. The propensity score condenses multivariate covariate information into a scalar to allow estimation of causal effects without specifying a model for how each covariate relates to the outcome. Avoiding specification of a detailed model for the outcome response surface is valuable for robust estimation of causal effects, but this strategy is at odds with the use of Bayes’ theorem, which presupposes a full probability model for the observed data that adheres to the likelihood principle. The goal of this article is to explicate this fundamental feature of Bayesian estimation of causal effects with propensity scores to provide context for the existing literature and for future work on this important topic.[Received June 2014. Revised September 2015.]
    Date Added 11/25/2019, 7:36:13 AM
    Modified 11/25/2019, 7:37:12 AM

    Tags:

    • bayes
    • causal-inference
    • propensity
    • causality
  • Bayesian propensity score analysis for observational data

    Type Journal Article
    Author Lawrence C. McCandless
    Author Paul Gustafson
    Author Peter C. Austin
    Volume 28
    Pages 94-112
    Publication Stat Med
    Date 2009
    Extra Citation Key: mcc09bay tex.citeulike-article-id= 13265723 tex.posted-at= 2014-07-14 14:10:02 tex.priority= 0
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • confounding
    • observational-study
    • propensity-score
    • bias
    • causal-inference
    • bayesian-statistics

    Notes:

    • using Bayesian credible intervals to adjust for uncertainty in estimation of propensity score;relied heavily on Rubin 5-category propensity adjustment

  • When should epidemiologic regressions use random coefficients?

    Type Journal Article
    Author Sander Greenland
    URL http://dx.doi.org/10.1111/j.0006-341X.2000.00915.x
    Volume 56
    Pages 915-921
    Publication Biometrics
    Date 2000
    Extra Citation Key: gre00whe tex.citeulike-article-id= 13265446 tex.citeulike-linkout-0= http://dx.doi.org/10.1111/j.0006-341X.2000.00915.x tex.posted-at= 2014-07-14 14:09:57 tex.priority= 0
    DOI 10.1111/j.0006-341X.2000.00915.x
    Date Added 7/7/2018, 1:38:33 PM
    Modified 11/8/2019, 8:01:59 AM

    Tags:

    • bayesian-methods
    • shrinkage
    • causal-inference
    • empirical-bayes-estimators
    • epidemiologic-method
    • hierarchical-regression
    • mixed-models
    • multilevel-modeling
    • random-coefficient-regression
    • variance-components

    Notes:

    • use of statistics in epidemiology is largely primitive;stepwise variable selection on confounders leaves important confounders uncontrolled;composition matrix;example with far too many significant predictors with many regression coefficients absurdly inflated when overfit;lack of evidence for dietary effects mediated through constituents;shrinkage instead of variable selection;larger effect on confidence interval width than on point estimates with variable selection;uncertainty about variance of random effects is just uncertainty about prior opinion;estimation of variance is pointless;instead the analysis should be repeated using different values;"if one feels compelled to estimate $\tau^{2}$, I would recommend giving it a proper prior concentrated amount contextually reasonable values";claim about ordinary MLE being unbiased is misleading because it assumes the model is correct and is the only model entertained;shrinkage towards compositional model;"models need to be complex to capture uncertainty about the relations...an honest uncertainty assessment requires parameters for all effects that we know may be present. This advice is implicit in an antiparsimony principle often attributed to L. J. Savage 'All models should be as big as an elephant (see Draper, 1995)'". See also gus06per.