Reblogging Dec. 31, 2011:
You know how in that recent movie, “Midnight in Paris,” the main character (I forget who plays it, I saw it on a plane) is a writer finishing a novel, and he steps into a cab that mysteriously picks him up at midnight and transports him back in time where he gets to run his work by such famous authors as Hemingway and Virginia Wolf? He is impressed when his work earns their approval and he comes back each night in the same mysterious cab…Well, imagine an error statistical philosopher is picked up in a mysterious taxi at midnight (New Year’s Eve
2011 2012) and is taken back fifty years and, lo and behold, finds herself in the company of Allan Birnbaum.[i]
ERROR STATISTICIAN: It’s wonderful to meet you Professor Birnbaum; I’ve always been extremely impressed with the important impact your work has had on philosophical foundations of statistics. I happen to be writing on your famous argument about the likelihood principle (LP). (whispers: I can’t believe this!)
BIRNBAUM: Ultimately you know I rejected the LP as failing to control the error probabilities needed for my Confidence concept.
ERROR STATISTICIAN: Yes, but I actually don’t think your argument shows that the LP follows from such frequentist concepts as sufficiency S and the weak conditionality principle WLP.[ii] Sorry,…I know it’s famous… Read more
In writing a new chapter on the Strong Likelihood Principle [i] the past few weeks, I noticed a passage in G. Casella and R. Berger (2002) that in turn recalled a puzzling remark noted in my Jan. 3, 2012 post. The post began:
A question arose from a Bayesian acquaintance:
“Although the Birnbaum result is of primary importance for sampling theorists, I’m still interested in it because many Bayesian statisticians think that model checking violates the (strong) likelihood principle (SLP), as if this principle is a fundamental axiom of Bayesian statistics”.
But this is puzzling for two reasons. First, if the LP does not preclude testing for assumptions (and he is right that it does not[ii]), then why not simply explain that rather than appeal to a disproof of something that actually never precluded model testing? To take the disproof of the LP as grounds to announce: “So there! Now even Bayesians are free to test their models” would seem only to ingrain the original fallacy.
You can read the rest of the original post here.
The remark in G. Casella and R. Berger seems to me equivocal on this point: Read more
*See also earlier posts from the CMU workshop here and here.
Elliott Sober has been writing on simplicity for a long time, so it was good to hear his latest thinking. If I understood him, he continues to endorse a comparative likelihoodist account, but he allows that, in model selection, “parsimony fights likelihood,” while, in adequate evolutionary theory, the two are thought to go hand in hand. Where it seems needed, therefore, he accepts a kind of “pluralism”. His discussion of the rival models in evolutionary theory and how they may give rise to competing likelihoods (for “tree taxonomies”) bears examination in its own right, but being in no position to accomplish this, I shall limit my remarks to the applicability of Sober’s insights (as my notes reflect them) to the philosophy of statistics and statistical evidence.
1. Comparativism: We can agree that a hypothesis is not appraised in isolation, but to say that appraisal is “contrastive” or “comparativist” is ambiguous. Error statisticians view hypothesis testing as between exhaustive hypotheses H and not-H (usually within a model), but deny that the most that can be said is that one hypothesis or model is comparatively better than another, among a group of hypotheses that is to be delineated at the outset. There’s an important difference here. The best-tested of the lot need not be well-tested!
2. Falsification: Sober made a point of saying that his account does not falsify models or hypotheses. We are to start out with all the possible models to be considered (hopefully including one that is true or approximately true), akin to the “closed universe” of standard Bayesian accounts[i], but do we not get rid of any as falsified, given data? It seems not.
APRIL FOOL’S DAY POST: This morning I received a paper I have been asked to review (anonymously as is typical). It is to head up a forthcoming issue of a new journal called Philosophy of Statistics: Retraction Watch. This is the first I’ve heard of the journal, and I plan to recommend they publish the piece, conditional on revisions. I thought I would post the abstract here. It’s that interesting.
“Some Slightly More Realistic Self-Criticism in Recent Work in Philosophy of Statistics,” Philosophy of Statistics: Retraction Watch, Vol. 1, No. 1 (2012), pp. 1-19.
In this paper we delineate some serious blunders that we and others have made in published work on frequentist statistical methods. First, although we have claimed repeatedly that a core thesis of the frequentist testing approach is that a hypothesis may be rejected with increasing confidence as the power of the test increases, we now see that this is completely backwards, and we regret that we have never addressed, or even fully read, the corrections found in Deborah Mayo’s work since at least 1983, and likely even before that.
Second, we have been wrong to claim that Neyman-Pearson (N-P) confidence intervals are inconsistent because in special cases it is possible for a specific 95% confidence interval to be known to be correct. Not only are the examples required to show this absurdly artificial, but the frequentist could simply interpret this “vacuous interval” “as a statement that all parameter values are consistent with the data at a particular level,” which, as Cox and Hinkley note, is an informative statement about the limitations in the data (Cox and Hinkley 1974, 226). Read more
Dear Reader: Not having been at this very long, I don’t know if it’s common for bloggers to collect a pile of rejected posts that one thinks better of before posting. Well, here’s one that belongs up in a “rejected post” page (and will be tucked away soon enough), but since we have so recently posted the Fisher-Neyman-Pearson “triad”, the blog-elders of Elba have twisted my elbow (repeatedly) to share this post, from back in the fall of 2011, London. Sincerely, D. Mayo
Egon Pearson on a Gate (by D. Mayo)
Did you ever consider how some of the colorful exchanges among better-known names in statistical foundations could be the basis for high literary drama in the form of one-act plays (even if appreciated by only 3-7 people in the world)? (Think of the expressionist exchange between Bohr and Heisenberg in Michael Frayn’s play Copenhagen, except here there would be no attempt at all to popularize—only published quotes and closely remembered conversations would be included, with no attempt to create a “story line”.) Somehow I didn’t think so. But rereading some of Savage’s high-flown praise of Birnbaum’s “breakthrough” argument (for the Likelihood Principle) today, I was swept into a “(statistical) theater of the absurd” mindset.
A Bayesian acquaintance writes:
Although the Birnbaum result is of primary importance for sampling theorists, I’m still interested in it because many Bayesian statisticians think that model checking violates the likelihood principle, as if this principle is a fundamental axiom of Bayesian statistics.
But this is puzzling for two reasons. First, if the LP does not preclude testing for assumptions (and he is right that it does not[i]), then why not simply explain that rather than appeal to a disproof of something that actually never precluded model testing? To take the disproof of the LP as grounds to announce: “So there! Now even Bayesians are free to test their models” would seem only to ingrain the original fallacy. Read more
You know how in that recent movie, “Midnight in Paris,” the main character (I forget who plays it, I saw it on a plane) is a writer finishing a novel, and he steps into a cab that mysteriously picks him up at midnight and transports him back in time where he gets to run his work by such famous authors as Hemingway and Virginia Wolf? He is impressed when his work earns their approval and he comes back each night in the same mysterious cab…Well, imagine an error statistical philosopher is picked up in a mysterious taxi at midnight (new Year’s Eve 2011) and is taken back fifty years and, lo and behold, finds herself in the company of Allan Birnbaum.[i] Read more
There is an often-heard slogan about the stages of the acceptance of novel truths:
First people deny a thing.
Then they belittle it.
Then they say they knew it all along.
I don’t know who was first to state it in one form or another. Here’s Schopenhauer with a slightly different variant:
“All truth passes through three stages: First, it is ridiculed; Second, it is violently opposed; and Third, it is accepted as self-evident.” – Arthur Schopenhauer
After recently presenting my paper criticizing the Birnbaum result on the likelihood principle (LP) the reception of my analysis seems somewhere around stage two, in some cases, moving into stage three (see my blogposts of December 6 and 7, 2011). Read more
Reblogging from a year ago. The Appendix of the “Cox/Mayo Conversation” (linked below [i]) is an attempt to quickly sketch Birnbaum’s argument for the strong likelihood principle (SLP), and its sins. Couple of notes: Firstly, I am a philosopher (of science and statistics) not a statistician. That means, my treatment will show all of the typical (and perhaps annoying) signs of being a trained philosopher-logician. I’ve no doubt statisticians would want to use different language, which is welcome. Second, this is just a blog (although perhaps my published version is still too informal for some). Read more