Archive for p-values

contemporary issues in hypothesis testing

Posted in Statistics with tags , , , , , , , , , , , , , , , , , , on September 26, 2016 by xi'an

hipocontemptThis week [at Warwick], among other things, I attended the CRiSM workshop on hypothesis testing, giving the same talk as at ISBA last June. There was a most interesting and unusual talk by Nick Chater (from Warwick) about the psychological aspects of hypothesis testing, namely about the unnatural features of an hypothesis in everyday life, i.e., how far this formalism stands from human psychological functioning.  Or what we know about it. And then my Warwick colleague Tom Nichols explained how his recent work on permutation tests for fMRIs, published in PNAS, testing hypotheses on what should be null if real data and getting a high rate of false positives, got the medical imaging community all up in arms due to over-simplified reports in the media questioning the validity of 15 years of research on fMRI and the related 40,000 papers! For instance, some of the headings questioned the entire research in the area. Or transformed a software bug missing the boundary effects into a major flaw.  (See this podcast on Not So Standard Deviations for a thoughtful discussion on the issue.) One conclusion of this story is to be wary of assertions when submitting a hot story to journals with a substantial non-scientific readership! The afternoon talks were equally exciting, with Andrew explaining to us live from New York why he hates hypothesis testing and prefers model building. With the birthday model as an example. And David Draper gave an encompassing talk about the distinctions between inference and decision, proposing a Jaynes information criterion and illustrating it on Mendel‘s historical [and massaged!] pea dataset. The next morning, Jim Berger gave an overview on the frequentist properties of the Bayes factor, with in particular a novel [to me] upper bound on the Bayes factor associated with a p-value (Sellke, Bayarri and Berger, 2001)

B¹⁰(p) ≤ 1/-e p log p

with the specificity that B¹⁰(p) is not testing the original hypothesis [problem] but a substitute where the null is the hypothesis that p is uniformly distributed, versus a non-parametric alternative that p is more concentrated near zero. This reminded me of our PNAS paper on the impact of summary statistics upon Bayes factors. And of some forgotten reference studying Bayesian inference based solely on the p-value… It is too bad I had to rush back to Paris, as this made me miss the last talks of this fantastic workshop centred on maybe the most important aspect of statistics!

Validity and the foundations of statistical inference

Posted in Statistics with tags , , , , , , , , on July 29, 2016 by xi'an

Natesh pointed out to me this recent arXival with a somewhat grandiose abstract:

In this paper, we argue that the primary goal of the foundations of statistics is to provide data analysts with a set of guiding principles that are guaranteed to lead to valid statistical inference. This leads to two new questions: “what is valid statistical inference?” and “do existing methods achieve this?” Towards answering these questions, this paper makes three contributions. First, we express statistical inference as a process of converting observations into degrees of belief, and we give a clear mathematical definition of what it means for statistical inference to be valid. Second, we evaluate existing approaches Bayesian and frequentist approaches relative to this definition and conclude that, in general, these fail to provide valid statistical inference. This motivates a new way of thinking, and our third contribution is a demonstration that the inferential model framework meets the proposed criteria for valid and prior-free statistical inference, thereby solving perhaps the most important unsolved problem in statistics.

Since solving the “most important unsolved problem in statistics” sounds worth pursuing, I went and checked the paper‘s contents.

“To us, the primary goal of the foundations of statistics is to provide a set of guiding principles that, if followed, will guarantee validity of the resulting inference. Our motivation for writing this paper is to be clear about what is meant by valid inference and to provide the necessary principles to help data analysts achieve validity.”

Which can be interpreted in so many ways that it is somewhat meaningless…

“…if real subjective prior information is available, we recommend using it. However, there is an expanding collection of work (e.g., machine learning, etc) that takes the perspective that no real prior information is available. Even a large part of the literature claiming to be Bayesian has abandoned the interpretation of the prior as a serious part of the model, opting for “default” prior that “works.” Our choice to omit a prior from the model is not for the (misleading) purpose of being “objective”—subjectivity is necessary—but, rather, for the purpose of exploring what can be done in cases where a fully satisfactory prior is not available, to see what improvements can be made over the status quo.”

This is a pretty traditional criticism of the Bayesian approach, namely that if a “true” prior is provided (by whom?) then it is optimal to use it. But this amounts to turn the prior into another piece of the sampling distribution and is not in my opinion a Bayesian argument! Most of the criticisms in the paper are directed at objective Bayes approaches, with the surprising conclusion that, because there exist cases where no matching prior is available, “the objective Bayesian approach [cannot] be considered as a general framework for scientific inference.” (p.9)

Another section argues that a Bayesian modelling cannot describe a state of total ignorance. This is formally correct, which is why there is no such thing as a non-informative or the non-informative prior, as often discussed here, but is this truly relevant, in that the inference problem contains one way or another information about the parameter, for instance through a loss function or a pseudo-likelihood.

“This is a desirable property that most existing methods lack.”

The proposal central to the paper thesis is to replace posterior probabilities by belief functions b(.|X), called statistical inference, that are interpreted as measures of evidence about subsets A of the parameter space. If not necessarily as probabilities. This is not very novel, witness the works of Dempster, Shafer and subsequent researchers. And not very much used outside Bayesian and fiducial statistics because of the mostly impossible task of defining a function over all subsets of the parameter space. Because of the subjectivity of such “beliefs”, they will be “valid” only if they are well-calibrated in the sense of b(A|X) being sub-uniform, that is, more concentrated near zero than a uniform variate (i.e., small) under the alternative, i.e. when θ is not in A. At this stage, since this is a mix of a minimax and proper coverage condition, my interest started to quickly wane… Especially because the sub-uniformity condition is highly demanding, if leading to controls over the Type I error and the frequentist coverage. As often, I wonder at the meaning of a calibration property obtained over all realisations of the random variable and all values of the parameter. So for me stability is neither “desirable” nor “essential”. Overall, I have increasing difficulties in perceiving proper coverage as a relevant property. Which has no stronger or weaker meaning that the coverage derived from a Bayesian construction.

“…frequentism does not provide any guidance for selecting a particular rule or procedure.”

I agree with this assessment, which means that there is no such thing as frequentist inference, but rather a philosophy for assessing procedures. That the Gleser-Hwang paradox invalidates this philosophy sounds a bit excessive, however. Especially when the bounded nature of Bayesian credible intervals is also analysed as a failure. A more relevant criticism is the lack of directives for picking procedures.

“…we are the first to recognize that the belief function’s properties are necessary in order for the inferential output to satisfy the required validity property”

The construction of the “inferential model” proposed by the authors offers similarities withn fiducial inference, in that it builds upon the representation of the observable X as X=a(θ,U). With further constraints on the function a() to ensure the validity condition holds… An interesting point is that the functional connection X=a(θ,U) means that the nature of U changes once X is observed, albeit in a delicate manner outside a Bayesian framework. When illustrated on the Gleser-Hwang paradox, the resolution proceeds from an arbitrary choice of a one-dimensional summary, though. (As I am reading the paper, I realise it builds on other and earlier papers by the authors, papers that I cannot read for lack of time. I must have listned to a talk by one of the authors last year at JSM as this rings a bell. Somewhat.) In conclusion of a quick Sunday afternoon read, I am not convinced by the arguments in the paper and even less by the impression of a remaining arbitrariness in setting the resulting procedure.

not an ASA’s statement on p-values

Posted in Books, Kids, Statistics, University life with tags , , , , on March 18, 2016 by xi'an

 

This may be a coincidence, but a few days after the ASA statement got published, Yuri Gurevich and Vladimir Vovk arXived a note on the Fundamentals of p-values. Which actually does not contribute to the debate. The paper is written in a Q&A manner. And defines a sort of peculiar logic related with [some] p-values. A second and more general paper is in the making, which may shed more light on the potential appeal of this formalism…

ASA’s statement on p-values [#2]

Posted in Books, Kids, Statistics, University life with tags , , , , , , , , on March 9, 2016 by xi'an

 

It took a visit on FiveThirtyEight to realise the ASA statement I mentioned yesterday was followed by individual entries from most members of the panel, much more diverse and deeper than the statement itself! Without discussing each and all comments, some points I subscribe to

  • it does not make sense to try to replace the p-value and the 5% boundary by something else but of the same nature. This was the main line of our criticism of Valen Johnson’s PNAS paper with Andrew.
  • it does not either make sense to try to come up with a hard set answer about whether or not a certain parameter satisfies a certain constraint. A comparison of predictive performances at or around the observed data sounds much more sensible, if less definitive.
  • the Bayes factor is often advanced as a viable alternative to the p-value in those comments, but it suffers from difficulties exposed in our recent testing by mixture paper, one being the lack of absolute scale.
  • we seem unable to escape the landscape set by Neyman and Pearson when constructing their testing formalism, including the highly unrealistic 0-1 loss function. And the grossly asymmetric opposition between null and alternative hypotheses.
  • the behaviour of any procedure of choice should be evaluated under different scenarios, most likely by simulation, including some accounting for misspecified models. Which may require an extra bit of non-parametrics. And we should abstain from considering further than evaluating whether or not the data looks compatible with each of the scenarios. Or how much through the mixture representation.

ASA’s statement on p-values

Posted in Books, Statistics, University life with tags , , , , , on March 8, 2016 by xi'an

 

Last night I received an email from the ASA signed by Jessica Utts and Ron Wasserstein with the following sentence

“Widespread use of ‘statistical significance’ (generally interpreted as ‘p

In short, we envision a new era, in which the broad scientific community recognizes what statisticians have been advocating for many years. In this “post p

Is such an era beyond reach? We think not, but we need your help in making sure this opportunity is not lost.”

which is obviously missing important bits. The email was pointing out a free access American Statistician article warning about the misuses and over-interpretations of p-values. Which contains rather basic “principles” that p-values are not probabilities that the null is true, that there is no golden level against which to compare the p-value, that nominal p-values may be far from actual p-values, that they do not provide a measure of evidence per se, &tc. As written in the conclusion, “Nothing in the ASA statement is new”. But, besides calling for caution and the cumulative use of different assessments of evidence, this statement may leave the non-statistician completely nonplussed about how to proceed when testing hypotheses or comparing models. And make the decision of Basic and Applied Social Psychology of rejecting all arguments based on p-values sound sensible.

Incidentally, the article contains the completion of the first sentence [in red below], if not of the second:

“Widespread use of ‘statistical significance’ (generally interpreted as ‘p≤ 0.05”) as a license for making a claim of a scientific finding (or implied truth) leads to considerable distortion of the scientific process.

 

It’s the selection’s fault not the p-values’… [seminar]

Posted in pictures, Statistics, University life with tags , , , , , , on February 5, 2016 by xi'an

Paris and la Seine, from Pont du Garigliano, Oct. 20, 2011Yoav Benjamini will give a seminar talk in Paris next Monday on the above (full title: “The replicability crisis in science: It’s the selection’s fault not the p-values’“). (That I will miss for being in Warwick at the time.) With a fairly terse abstract:

I shall discuss the problem of lack of replicability of results in science, and point at selective inference as a statistical root cause. I shall then present a few strategies for addressing selective inference, and their application in genomics, brain research and earlier phases of clinical trials where both primary and secondary endpoints are being used.

Details: February 8, 2016, 16h, Université Pierre & Marie Curie, campus Jussieu, salle 15-16-101.

how individualistic should statistics be?

Posted in Books, pictures, Statistics with tags , , , , , , , , , , , on November 5, 2015 by xi'an

keep-stats-subjectiveKeli Liu and Xiao-Li Meng completed a paper on the very nature of inference, to appear in The Annual Review of Statistics and Its Application. This paper or chapter is addressing a fundamental (and foundational) question on drawing inference based a sample on a new observation. That is, in making prediction. To what extent should the characteristics of the sample used for that prediction resemble those of the future observation? In his 1921 book, A Treatise on Probability, Keynes thought this similarity (or individualisation) should be pushed to its extreme, which led him to somewhat conclude on the impossibility of statistics and never to return to the field again. Certainly missing the incoming possibility of comparing models and selecting variables. And not building so much on the “all models are wrong” tenet. On the contrary, classical statistics use the entire data available and the associated model to run the prediction, including Bayesian statistics, although it is less clear how to distinguish between data and control there. Liu & Meng debate about the possibility of creating controls from the data alone. Or “alone” as the model behind always plays a capital role.

“Bayes and Frequentism are two ends of the same spectrum—a spectrum defined in terms of relevance and robustness. The nominal contrast between them (…) is a red herring.”

viemortrerbThe paper makes for an exhilarating if definitely challenging read. With a highly witty writing style. If only because the perspective is unusual, to say the least!, and requires constant mental contortions to frame the assertions into more traditional terms.  For instance, I first thought that Bayesian procedures were in agreement with the ultimate conditioning approach, since it conditions on the observables and nothing else (except for the model!). Upon reflection, I am not so convinced that there is such a difference with the frequentist approach in the (specific) sense that they both take advantage of the entire dataset. Either from the predictive or from the plug-in distribution. It all boils down to how one defines “control”.

“Probability and randomness, so tightly yoked in our minds, are in fact distinct concepts (…) at the end of the day, probability is essentially a tool for bookkeeping, just like the abacus.”

Some sentences from the paper made me think of ABC, even though I am not trying to bring everything back to ABC!, as drawing controls is the nature of the ABC game. ABC draws samples or control from the prior predictive and only keeps those for which the relevant aspects (or the summary statistics) agree with those of the observed data. Which opens similar questions about the validity and precision of the resulting inference, as well as the loss of information due to the projection over the summary statistics. While ABC is not mentioned in the paper, it can be used as a benchmark to walk through it.

“In the words of Jack Kiefer, we need to distinguish those problems with `luck data’ from those with `unlucky data’.”

keep-calm-and-condi-tionI liked very much recalling discussions we had with George Casella and Costas Goutis in Cornell about frequentist conditional inference, with the memory of Jack Kiefer still lingering around. However, I am not so excited about the processing of models here since, from what I understand in the paper (!), the probabilistic model behind the statistical analysis must be used to some extent in producing the control case and thus cannot be truly assessed with a critical eye. For instance, of which use is the mean square error when the model behind is unable to produce the observed data? In particular, the variability of this mean squared error is directly driven by this model. Similarly the notion of ancillaries is completely model-dependent. In the classification diagrams opposing robustness to relevance, all methods included therein are parametric. While non-parametric types of inference could provide a reference or a calibration ruler, at the very least.

Also, by continuously and maybe a wee bit heavily referring to the doctor-and-patient analogy, the paper is somewhat confusing as to which parts are analogy and which parts are methodology and to which type of statistical problem is covered by the discussion (sometimes it feels like all problems and sometimes like medical trials).

“The need to deliver individualized assessments of uncertainty are more pressing than ever.”

 A final question leads us to an infinite regress: if the statistician needs to turn to individualized inference, at which level of individuality should the statistician be assessed? And who is going to provide the controls then? In any case, this challenging paper is definitely worth reading by (only mature?) statisticians to ponder about the nature of the game!