RMM-8: New Mayo paper: “StatSci and PhilSci: part 2 (Shallow vs Deep Explorations)”

A new article of mine,  “Statistical Science and Philosophy of Science Part 2: Shallow versus Deep Explorations” has been published in the on-line journal, Rationality, Markets, and Morals (Special Topic: Statistical Science and Philosophy of Science: Where Do/Should They Meet?”).

The contributions to this special volume began with the conference we ran in June 2010. (See web poster.)   My first article in this collection was essentially just my introduction to the volume, whereas this new one discusses my work. If you are a reader of this blog, you will recognize portions from early posts, as I’d been revising it then.

The sections are listed below. I will be posting portions in the next few days. We invite comments for this blog, and for possible publication in this special volume of RMM, if received before the end of this year.

This is the 8th RMM announcement. Many thanks to Sailor for digging up the previous 7, and listing them at the end*. (The paper’s title stemmed from the Deepwater Horizon oil spill of spring 2010**).


Inability to clearly defend against the criticisms of frequentist methods has turned many a frequentist away from venturing into foundational battlegrounds. Conceding the distorted perspectives drawn from overly literal and radical expositions of what Fisher, Neyman, and Pearson ‘really thought’, some deny they matter to current practice. The goal of this paper is not merely to call attention to the howlers that pass as legitimate criticisms of frequentist error statistics, but also to sketch the main lines of an alternative statistical philosophy within which to better articulate the roles and value of frequentist tools.

Statistical Science Meets Philosophy of Science Part 2:
Shallow versus Deep Explorations

 1. Comedy Hour at the Bayesian Retreat

2. Popperians Are to Frequentists as Carnapians Are to Bayesians
2.1 Severe Tests
2.2 Another Egregious Violation of the Severity Requirement
2.3 The Rationale for Severity is to Find Things Out Reliably
2.4 What Can Be Learned from Popper; What Can Popperians Be Taught?

3. Frequentist Error-Statistical Tests
3.1 Probability in Statistical Models of Experiments
3.2 Statistical Test Ingredients
3.3. Hypotheses and Events
3.4. Hypotheses Inferred Need Not Be Predesignated

4. Neyman’s Inferential Side: Neyman on Carnap
4.1 Frequentist Statistics Is Not the Frequentist ‘Straight Rule’
4.2 Post-Data Uses of Power
4.3 One-sided Test T+
4.4 Frequentist Principle of Evidence: FEV
4.5 Pragmatism without Subjectivism

5. The Error-Statistical Philosophy
5.1 Error (Probability) Statistics
5.2 Philosophy-Laden Criticisms of Frequentist Statistical Methods
5.3 Severity as a ‘Metastatistical’ Assessment

6. Some Knock-Down Criticisms of Frequentist Error Statistics
6.1 Fallacies of Rejection
6.2 P-values Conflict with Posterior Probabilities: The Criticism in Statistics
6.3 Severity Values Conflict with Posteriors: The Criticism in Philosophy
6.4 Trivial Intervals and Allegations of Unsoundness
6.5 Getting Credit (or Blamed) for Something You Didn’t Do
6.6 Two Measuring Instruments with Different Precisions

7. Can/Should Bayesian and Error Statistical Philosophies Be Reconciled?
7.1 The (Strong) Likelihood Principle (LP)
7.2 Optional Stopping Effect
7.3 The Optional Stopping Effect with (Two-sided) Confidence Intervals
7.4 Savage’s Sleight of Hand in Defense of the LP
7.5 The Counterrevolution?

8. Concluding Remarks: Deep versus Shallow Statistical Philosophy


*Previous RMM article announcements:

*The “deep exploration” metaphor grew out of the (Gulf of Mexico) Deepwater Horizon oil spill of April -July 2010. I found the whole episode fascinating, what with blowout preventers, blind sheer rams, and deep ocean engineering robots.

PhilStock: The stock, Diamond Offshore (DO), which I’d long owned, dropped dramatically even though it had nothing to do with it.  Leases were not renewed and some of its ships had to depart for Brazil. However, it has a large special dividend (you will not see it in the regular dividend), and it’s near its yearly high today (but earnings are tomorrow).  DO is the stock mascot of this blog (whatever that means).

Categories: Philosophy of Statistics, Statistics | Tags: ,

Post navigation

Comments are closed.

Blog at WordPress.com.