## the maths of Jeffreys-Lindley paradox

Posted in Books, Kids, Statistics with tags , , , , , , , on March 26, 2015 by xi'an

Cristiano Villa and Stephen Walker arXived on last Friday a paper entitled On the mathematics of the Jeffreys-Lindley paradox. Following the philosophical papers of last year, by Ari Spanos, Jan Sprenger, Guillaume Rochefort-Maranda, and myself, this provides a more statistical view on the paradox. Or “paradox”… Even though I strongly disagree with the conclusion, namely that a finite (prior) variance σ² should be used in the Gaussian prior. And fall back on classical Type I and Type II errors. So, in that sense, the authors avoid the Jeffreys-Lindley paradox altogether!

The argument against considering a limiting value for the posterior probability is that it converges to 0, 21, or an intermediate value. In the first two cases it is useless. In the medium case. achieved when the prior probability of the null and alternative hypotheses depend on variance σ². While I do not want to argue in favour of my 1993 solution

$\rho(\sigma) = 1\big/ 1+\sqrt{2\pi}\sigma$

since it is ill-defined in measure theoretic terms, I do not buy the coherence argument that, since this prior probability converges to zero when σ² goes to infinity, the posterior probability should also go to zero. In the limit, probabilistic reasoning fails since the prior under the alternative is a measure not a probability distribution… We should thus abstain from over-interpreting improper priors. (A sin sometimes committed by Jeffreys himself in his book!)

## the demise of the Bayes factor

Posted in Books, Kids, Statistics, Travel, University life with tags , , , , , , , , on December 8, 2014 by xi'an

With Kaniav Kamary, Kerrie Mengersen, and Judith Rousseau, we have just arXived (and submitted) a paper entitled “Testing hypotheses via a mixture model”. (We actually presented some earlier version of this work in Cancũn, Vienna, and Gainesville, so you may have heard of it already.) The notion we advocate in this paper is to replace the posterior probability of a model or an hypothesis with the posterior distribution of the weights of a mixture of the models under comparison. That is, given two models under comparison,

$\mathfrak{M}_1:x\sim f_1(x|\theta_1) \text{ versus } \mathfrak{M}_2:x\sim f_2(x|\theta_2)$

we propose to estimate the (artificial) mixture model

$\mathfrak{M}_{\alpha}:x\sim\alpha f_1(x|\theta_1) + (1-\alpha) f_2(x|\theta_2)$

and in particular derive the posterior distribution of α. One may object that the mixture model is neither of the two models under comparison but this is the case at the boundary, i.e., when α=0,1. Thus, if we use prior distributions on α that favour the neighbourhoods of 0 and 1, we should be able to see the posterior concentrate near 0 or 1, depending on which model is true. And indeed this is the case: for any given Beta prior on α, we observe a higher and higher concentration at the right boundary as the sample size increases. And establish a convergence result to this effect. Furthermore, the mixture approach offers numerous advantages, among which [verbatim from the paper]:

## reliable ABC model choice via random forests

Posted in pictures, R, Statistics, University life with tags , , , , , , , on October 29, 2014 by xi'an

After a somewhat prolonged labour (!), we have at last completed our paper on ABC model choice with random forests and submitted it to PNAS for possible publication. While the paper is entirely methodological, the primary domain of application of ABC model choice methods remains population genetics and the diffusion of this new methodology to the users is thus more likely via a media like PNAS than via a machine learning or statistics journal.

When compared with our recent update of the arXived paper, there is not much different in contents, as it is mostly an issue of fitting the PNAS publication canons. (Which makes the paper less readable in the posted version [in my opinion!] as it needs to fit the main document within the compulsory six pages, relegated part of the experiments and of the explanations to the Supplementary Information section.)

## all models are wrong

Posted in Statistics, University life with tags , , , , , , , on September 27, 2014 by xi'an

“Using ABC to evaluate competing models has various hazards and comes with recommended precautions (Robert et al. 2011), and unsurprisingly, many if not most researchers have a healthy scepticism as these tools continue to mature.”

Michael Hickerson just published an open-access letter with the above title in Molecular Ecology. (As in several earlier papers, incl. the (in)famous ones by Templeton, Hickerson confuses running an ABC algorithm with conducting Bayesian model comparison, but this is not the main point of this post.)

“Rather than using ABC with weighted model averaging to obtain the three corresponding posterior model probabilities while allowing for the handful of model parameters (θ, τ, γ, Μ) to be estimated under each model conditioned on each model’s posterior probability, these three models are sliced up into 143 ‘submodels’ according to various parameter ranges.”

The letter is in fact a supporting argument for the earlier paper of Pelletier and Carstens (2014, Molecular Ecology) which conducted the above splitting experiment. I could not read this paper so cannot judge of the relevance of splitting this way the parameter range. From what I understand it amounts to using mutually exclusive priors by using different supports.

“Specifically, they demonstrate that as greater numbers of the 143 sub-models are evaluated, the inference from their ABC model choice procedure becomes increasingly.”

An interestingly cut sentence. Increasingly unreliable? mediocre? weak?

“…with greater numbers of models being compared, the most probable models are assigned diminishing levels of posterior probability. This is an expected result…”

True, if the number of models under consideration increases, under a uniform prior over model indices, the posterior probability of a given model mechanically decreases. But the pairwise Bayes factors should not be impacted by the number of models under comparison and the letter by Hickerson states that Pelletier and Carstens found the opposite:

“…pairwise Bayes factor[s] will always be more conservative except in cases when the posterior probabilities are equal for all models that are less probable than the most probable model.”

Which means that the “Bayes factor” in this study is computed as the ratio of a marginal likelihood and of a compound (or super-marginal) likelihood, averaged over all models and hence incorporating the prior probabilities of the model indices as well. I had never encountered such a proposal before. Contrary to the letter’s claim:

“…using the Bayes factor, incorporating all models is perhaps more consistent with the Bayesian approach of incorporating all uncertainty associated with the ABC model choice procedure.”

Besides the needless inclusion of ABC in this sentence, a somewhat confusing sentence, as Bayes factors are not, stricto sensu, Bayesian procedures since they remove the prior probabilities from the picture.

“Although the outcome of model comparison with ABC or other similar likelihood-based methods will always be dependent on the composition of the model set, and parameter estimates will only be as good as the models that are used, model-based inference provides a number of benefits.”

All models are wrong but the very fact that they are models allows for producing pseudo-data from those models and for checking if the pseudo-data is similar enough to the observed data. In components that matters the most for the experimenter. Hence a loss function of sorts…

## ABC model choice by random forests [guest post]

Posted in pictures, R, Statistics, University life with tags , , , , , , , , , , on August 11, 2014 by xi'an

[Dennis Prangle sent me his comments on our ABC model choice by random forests paper. Here they are! And I appreciate very much contributors commenting on my paper or others, so please feel free to join.]

This paper proposes a new approach to likelihood-free model choice based on random forest classifiers. These are fit to simulated model/data pairs and then run on the observed data to produce a predicted model. A novel “posterior predictive error rate” is proposed to quantify the degree of uncertainty placed on this prediction. Another interesting use of this is to tune the threshold of the standard ABC rejection approach, which is outperformed by random forests.

The paper has lots of thought-provoking new ideas and was an enjoyable read, as well as giving me the encouragement I needed to read another chapter of the indispensable Elements of Statistical Learning However I’m not fully convinced by the approach yet for a few reasons which are below along with other comments.

Alternative schemes

The paper shows that random forests outperform rejection based ABC. I’d like to see a comparison to more efficient ABC model choice algorithms such as that of Toni et al 2009. Also I’d like to see if the output of random forests could be used as summary statistics within ABC rather than as a separate inference method.

Posterior predictive error rate (PPER)

This is proposed to quantify the performance of a classifier given a particular data set. The PPER is the proportion of times the classifier’s most favoured model is incorrect for simulated model/data pairs drawn from an approximation to the posterior predictive. The approximation is produced by a standard ABC analysis.

Misclassification could be due to (a) a poor classifier or (b) uninformative data, so the PPER aggregrates these two sources of uncertainty. I think it is still very desirable to have an estimate of the uncertainty due to (b) only i.e. a posterior weight estimate. However the PPER is useful. Firstly end users may sometimes only care about the aggregated uncertainty. Secondly relative PPER values for a fixed dataset are a useful measure of uncertainty due to (a), for example in tuning the ABC threshold. Finally, one drawback of the PPER is the dependence on an ABC estimate of the posterior: how robust are the results to the details of how this is obtained?

Classification

This paper illustrates an important link between ABC and machine learning classification methods: model choice can be viewed as a classification problem. There are some other links: some classifiers make good model choice summary statistics (Prangle et al 2014) or good estimates of ABC-MCMC acceptance ratios for parameter inference problems (Pham et al 2014). So the good performance random forests makes them seem a generally useful tool for ABC (indeed they are used in the Pham et al al paper).

## ABC model choice by random forests

Posted in pictures, R, Statistics, Travel, University life with tags , , , , , , , , , , , , , on June 25, 2014 by xi'an

After more than a year of collaboration, meetings, simulations, delays, switches,  visits, more delays, more simulations, discussions, and a final marathon wrapping day last Friday, Jean-Michel Marin, Pierre Pudlo,  and I at last completed our latest collaboration on ABC, with the central arguments that (a) using random forests is a good tool for choosing the most appropriate model and (b) evaluating the posterior misclassification error rather than the posterior probability of a model is an appropriate paradigm shift. The paper has been co-signed with our population genetics colleagues, Jean-Marie Cornuet and Arnaud Estoup, as they provided helpful advice on the tools and on the genetic illustrations and as they plan to include those new tools in their future analyses and DIYABC software.  ABC model choice via random forests is now arXived and very soon to be submitted…

One scientific reason for this fairly long conception is that it took us several iterations to understand the intrinsic nature of the random forest tool and how it could be most naturally embedded in ABC schemes. We first imagined it as a filter from a set of summary statistics to a subset of significant statistics (hence the automated ABC advertised in some of my past or future talks!), with the additional appeal of an associated distance induced by the forest. However, we later realised that (a) further ABC steps were counterproductive once the model was selected by the random forest and (b) including more summary statistics was always beneficial to the performances of the forest and (c) the connections between (i) the true posterior probability of a model, (ii) the ABC version of this probability, (iii) the random forest version of the above, were at best very loose. The above picture is taken from the paper: it shows how the true and the ABC probabilities (do not) relate in the example of an MA(q) model… We thus had another round of discussions and experiments before deciding the unthinkable, namely to give up the attempts to approximate the posterior probability in this setting and to come up with another assessment of the uncertainty associated with the decision. This led us to propose to compute a posterior predictive error as the error assessment for ABC model choice. This is mostly a classification error but (a) it is based on the ABC posterior distribution rather than on the prior and (b) it does not require extra-computations when compared with other empirical measures such as cross-validation, while avoiding the sin of using the data twice!

## Approximate Bayesian model choice

Posted in Books, R, Statistics, Travel, University life with tags , , , , , , , , , on March 17, 2014 by xi'an

The above is the running head of the arXived paper with full title “Implications of  uniformly distributed, empirically informed priors for phylogeographical model selection: A reply to Hickerson et al.” by Oaks, Linkem and Sukuraman. That I (again) read in the plane to Montréal (third one in this series!, and last because I also watched the Japanese psycho-thriller Midsummer’s Equation featuring a physicist turned detective in one of many TV episodes. I just found some common features with The Devotion of Suspect X, only to discover now that the book has been turned into another episode in the series.)

“Here we demonstrate that the approach of Hickerson et al. (2014) is dangerous in the sense that the empirically-derived priors often exclude from consideration the true values of the models’ parameters. On a more fundamental level, we question the value of adopting an empirical Bayesian stance for this model-choice problem, because it can mislead model posterior probabilities, which are inherently measures of belief in the models after prior knowledge is updated by the data.”

This paper actually is a reply to Hickerson et al. (2014, Evolution), which is itself a reply to an earlier paper by Oaks et al. (2013, Evolution). [Warning: I did not check those earlier references!] The authors object to the use of “narrow, empirically informed uniform priors” for the reason reproduced in the above quote. In connection with the msBayes of Huang et al. (2011, BMC Bioinformatics). The discussion is less about ABC used for model choice and posterior probabilities of models and more about the impact of vague priors, Oaks et al. (2013) arguing that this leads to a bias towards models with less parameters, a “statistical issue” in their words, while Hickerson et al. (2014) think this is due to msBayes way of selecting models and their parameters at random.

“…it is difficult to choose a uniformly distributed prior on divergence times that is broad enough to confidently contain the true values of parameters while being narrow enough to avoid spurious support of models with less parameter space.”

So quite an interesting debate that takes us in fine far away from the usual worries about ABC model choice! We are more at the level empirical versus natural Bayes, seen in the literature of the 80’s. (The meaning of empirical Bayes is not that clear in the early pages as the authors seem to involve any method using the data “twice”.) I actually do not remember reading papers about the formal properties of model choice done through classical empirical Bayes techniques. Except the special case of Aitkin’s (1991,2009) integrated likelihood. Which is essentially the analysis performed on the coin toy example (p.7)

“…models with more divergence parameters will be forced to integrate over much greater parameter space, all with equal prior density, and much of it with low likelihood.”

The above argument is an interesting rephrasing of Lindley’s paradox, which I cannot dispute, but of course it does not solve the fundamental issue of how to choose the prior away from vague uniform priors… I also like the quote “the estimated posterior probability of a model is a single value (rather than a distribution) lacking a measure of posterior uncertainty” as this is an issue on which we are currently working. I fully agree with the statement and we think an alternative assessment to posterior probabilities could be more appropriate for model selection in ABC settings (paper soon to come, hopefully!).