Asymptotic Distribution of P Values in Composite Null Models

By James M. Robins, A. van der Vaart, Valérie Ventura. 2000

255 citations recorded by the source corpus.

Subfields: Applied Mathematics, Artificial Intelligence, Mathematical Physics

View Paper: https://www.semanticscholar.org/paper/380f7bf31b898ace68caa597340e39e65926a87e

How researchers see this paper

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

In Applied Mathematics

In Artificial Intelligence

In Mathematical Physics

Citations in the corpus

Philosophy and the practice of Bayesian statistics (2010, unknown): “…checks have too little power to be useful statistical tools (Bayarri and Berger, 2000, 2004), how they might be modified to increase their power (Robins et al., 2000; Fraser and Rousseau, 2008), whether some form of empirical prior predictive check might not be better (Bayarri and Castellanos,…”
Type S error rates for classical and Bayesian single and multiple comparison procedures (2000, background): “Our answer is that Type 1 error calculations are important for understanding p-values for testing a null hypothesis, both in classical and Bayesian settings (Meng, 1994, Robins et al., 1998).”
Assessing Mediational Models: Testing and Interval Estimation for Indirect Effects (2010, background): “This helps remove (partial) the association between the indirect effect and the “nuisance” parameter (e.g., “).5 Asymptotically, pppost provides the correct probability of the observed data under the null hypothesis (Bayarri & Berger, 1999, 2000; Robins et al., 2000).”
Two simple examples for understanding posterior p-values whose distributions are far from uniform (2013, background): “…an applied direction, some researchers have proposed adjusted predictive checks that are calibrated to have asymptotic uniform null distributions (Robins, Vaart, and Ventura, 2000); others have argued that, in applied examples, posterior predictive checks are directly interpretable without the…”
Statistical Model Criticism using Kernel Two Sample Tests (2015, background): “Robins, van der Vaart & Ventura [28] demonstrated that Bayesian and plug-in p-values are not classical p-values (frequentist p-values in their terminology) i.e. they do not have a uniform distribution under the relevant null hypothesis.”
Metrics for External Model Evaluation with an Application to the Population Pharmacokinetics of Gliclazide (2006, background): “Yano, Beal and Sheiner defined posterior predictive check (PPC) (10) and proposed three approaches to compute the posterior distribution of the parameters estimated through the maximum likelihood estimation (31,32).”
On tests of spatial pattern based on simulation envelopes (2014, methodology): “The correct handling of P values for composite null hypotheses is regarded as an unresolved research problem in statistical inference (Brooks et al. 1997, Bayarri and Berger 2000, and Robins et al. 2000).”
Inferential Statistics as Descriptive Statistics: There Is No Replication Crisis if We Don’t Expect Replication (2018, methodology): “In contrast, Fisherian P -values can be used to test this assumption before using it in the Bayesian deduction (Bayarri and Berger 2000; Robins et al. 2000).”