Frequentists have long been in a kind of exile when it comes to statistical philosophy. The line is (or was) that only personalistic Bayesianism had a shot at respectable philosophical foundations. This may now be changing. Perhaps frequentist foundations, never made fully explicit, but at most lying deep below the ocean floor, are finally being disinterred. Join me, if you will, for a little deep-water drilling, as I cast about on my isle of Elba. (See Elba grease.)
This blog focuses on issues connecting statistical science and philosophy of science, philosophical debates about the nature of inductive inference and the roles of probability and statistical models in learning about phenomena in the face of limited information and threats of error. The blog has discussed ways that philosophical clarifications of central issues —objective/subjective, deduction/induction, and truth/idealizations— can illuminate foundational issues that occur within statistical practice. Central issues include fallacies of statistical tests, behavioristic vs. inferential construals of frequentist methods, the role of background knowledge in linking statistical and substantive inference, validating statistical models, reforms and reinterpretations of methods, and contemporary as well as historical attempts to reconcile or unify frequentist and Bayesian statistics. Links to statistics in the law, medicine, and economics are made at times with guest bloggers. The blog regularly has open discussions and invitations to contribute posts (“U-Phils”), and to forge interlinks with related blogs. Open forums have also included relevant papers by historical figures in statistics and philosophy, and special topics arising from conferences (recently, machine learning; ontology and methodology).
You can write to me at error@vt.edu
Error Statistics Philosophy 