# Statisfaction

## Turing revisited in Turin, and Oxford

Posted in General by Julyan Arbel on 18 June 2015

Are you paying attention? Good. If you are not listening carefully, you will miss things. Important things.

With colleagues Stefano Favaro and Bernardo Nipoti from Turin and Yee Whye Teh from Oxford, we have just arXived an article on discovery probabilities. If you are looking for some info on a space shuttle, a cycling team or a TV channel, it’s the wrong place. Instead, discovery probabilities are central to ecology, biology and genomics where data can be seen as a population of individuals belonging to an (ideally) infinite number of species. Given a sample of size $n$, the $l$-discovery probability $D_{n}(l)$ is the probability that the next individual observed matches a species with frequency $l$ in the $n$-sample. For instance, the probability of observing a new species $D_{n}(0)$ is key for devising sampling experiments.

By the way, why Alan Turing? Because with his fellow researcher at Bletchley Park Irving John Good, starred in The Imitation Game too, Turing is also known for the so-called Good-Turing estimator of the discovery probability

$(l+1)\frac{m_{l+1,n}}{n}$

which involves $m_{l+1,n}$, the number of species with frequency $l+1$ in the sample (ie frequencies frequency, if you follow me). As it happens, this estimator defined in Good 1953 Biometrika paper became wildly popular among ecology-biology-genomics communities since then, at least in the small circles where wild popularity and probability aren’t mutually exclusive.

Simple explicit estimators $\hat{\mathcal{D}}_{n}(l)$ of discovery probabilities in the Bayesian nonparametric (BNP) framework of Gibbs-type priors were given by Lijoi, Mena and Prünster in a 2007 Biometrika paper. The main difference between the two estimators of $D_{n}(l)$ is that Good-Turing involves $n$ and $m_{l+1,n}$ only, while the BNP involves $n$, $m_{l,n}$ (instead of $m_{l+1,n}$), and $k_n$, the total number of observed species. It has been shown in the literature that the BNP estimators are more reliable than Good-Turing estimators.

How do we contribute? (i) we describe the posterior distribution of the discovery probabilities in the BNP model, which is pretty useful for deriving exact credible intervals of the estimates, and (ii) we investigate large $n$ asymptotic behavior of the estimators.

## Who is Julia ?

Posted in General by JB Salomond on 4 June 2015

Hi there !

Unfortunately this post is indeed about statistics…

If you are randomly walking around the statistics blogs, you probably have certainly heard of this new language called Julia. It is said by the developers to be as easy to write as R and as fast as C (!) which is quite a catchy way of selling their work. After talking with a Julia enthusiastic user in Amsterdam, I decided to give it a try. And here I am sharing my first impressions.

Fist thing first, the installation is as easy as any other language, plus there is a neat Package management that allows you to get started quite easily. In this respect it is very similar to R.
On the minus side I became a big fan of RStudio Julian (… oupsy Julyan) told you about a long time ago. These kind of programs really make your life easier. I thus tried Juno which turned out to be cumbersome and terribly slow. I would have loved to have an IDE for Julia that would be up to the RStudio standard. Nevermind.

No lets talk a little about what is really interesting : “Is their catch phrase false advertising or not?!”.

There is a bunch of relatively good tutorials online which are really helpful to learn the basic vocabulary, but indeed if like me you are use to code in R and/or Python, you should get it pretty fast and can almost copy-paste your favourite code into Julia and with a few adjustments, it will work. So as easy to write as R : quite so.

I then tried to compare computational times for some of my latest codes and there came the good surprise ! A code that would take a handful of minutes to run in R mainly due to unavoidable loops took a couple of seconds to run in Julia, without any other sorts of optimization. The handling of big objects is smooth and I did not ran into memory problems that R was suffering from.

So far so good ! But of course there has to be some drawbacks. The first one is the poor package repository compare to CRAN or even what you can get for Python. This might of course improve in the next few years as the language is still quite new. However, it is bothering to have to re-code something when you are used to simply load a package in R. Another, probably less important problem, is the lack of data visualization methods and especially the absence of ggplot2 that we have grown quite found of around here. There is of course Gadfly, which is quite close but once again, it is up to now very limited compared to what I was used to…

All in all, I am happy to have tried Julia, and I am quite sure that I will be using it quite a lot from now on. However, even if from a efficiency point of view, it is great, and it is way easier to learn than C (which I should have done a while ago), R and its tremendous package repository is far from beaten.

Oh and by the way, it uses PyPlot based on MatplotLib that allow you to make some xkcd-like plots, which can make your presentations a lot more fun.

## Reading Bayesian classics — presentations

Posted in General by Julyan Arbel on 21 April 2015

The students did a great job in presenting some Bayesian classics. I enjoyed reading the papers (pdfs can be found here), most of which I hadn’t read before, and enjoyed also the students’ talks. I share here some of the best ones, as well as some demonstrative excerpts from the papers. In chronological order (presentations on slideshare below):

• W. Keith Hastings. Monte Carlo sampling methods using Markov chains and their applications. Biometrika, 57(1):97–109, 1970.

In this paper, we shall consider Markov chain methods of sampling that are generalizations of a method proposed by Metropolis et al. (1953), which has been used extensively for numerical problems in statistical mechanics.

• Dennis V. Lindley and Adrian F.M. Smith. Bayes estimates for the linear model. Journal of the Royal Statistical Society: Series B (Statistical Methodology), with discussion, 1–41, 1972.

From Prof. B. de Finetti discussion (note the valliant collaborator Smith!):

I think that the main point to stress about this interesting and important paper is its significance for the philosophical questions underlying the acceptance of the Bayesian standpoint as the true foundation for inductive reasoning, and in particular for statistical inference. So far as I can remember, the present paper is the first to emphasize the role of the Bayesian standpoint as a logical framework for the analysis of intricate statistical situation. […] I would like to express my warmest congratulations to my friend Lindley and his valiant collaborator Smith.

## Statistics journals network

Posted in General, R, Statistics by Julyan Arbel on 16 April 2015

Statistical journals friendship (clic for SVG format)

Xian blogged recently on the incoming RSS read paper: Statistical Modelling of Citation Exchange Between Statistics Journals, by Cristiano Varin, Manuela Cattelan and David Firth. Following the last JRSS B read paper by one of us! The data that are used in the paper (and can be downloaded here) are quite fascinating for us, academics fascinated by academic rankings, for better or for worse (ironic here). They consist in cross citations counts $C = (C_{ij})$ for 47 statistics journals (see list and abbreviations page 5): $C_{ij}$ is the number of citations from articles published in journal $j$ in 2010 to papers published in journal $i$ in the 2001-2010 decade. The choice of the list of journals is discussed in the paper. Major journals missing include Bayesian Analysis (published from 2006), The Annals of Applied Statistics (published from 2007).

I looked at the ratio of Total Citations Received by Total Citations made. This is a super simple descriptive statistic which happen to look rather similar to Figure 4 which plots Export Scores from Stigler model (can’t say more about it, I haven’t read in detail). The top five is the same modulo the swap between Annals of Statistics and Biometrika. Of course a big difference is that the Cited/Citation ratio isn’t endowed with a measure of uncertainty (below, left is my making, right is Fig. 4 in the paper).

I was surprised not to see a graph / network representation of the data in the paper. As it happens I wanted to try the gephi software for drawing graphs, used for instance by François Caron and Emily Fox in their sparse graphs paper. I got the above graph, where:

• for the data, I used the citations matrix $C$ renormalized by the total number of citations made, which I denote by $\tilde C$. This is a way to account for the size (number of papers published) of the journal. This is just a proxy though since the actual number of papers published by the journal is not available in the data. Without that correction, CSDA is way ahead of all the others.
• the node size represents the Cited/Citing ratio
• the edge width represents the renormalized $\tilde C_{ij}$. I’m unsure of what gephi does here, since it converts my directed graph into an undirected graph. I suppose that it displays only the largest of the two edges $\tilde C_{ij}$ and $\tilde C_{ji}$.
• for a better visibility I kept only the first decile of heaviest edges.
• the clusters identified by four colors are modularity classes obtained by the Louvain method.

Some remarks

The two software journals included in the dataset are quite outliers:

• the Journal of Statistical Software (JSS) is disconnected from the others, meaning it has no normalized citations $\tilde C_{ij}$ in the first decile. Except from its self citations which are quite big and make it the 4th Impact Factor from the total list in 2010 (and apparently the first in 2015).
• the largest $\tilde C_{ij}$ is the self citations of the STATA Journal (StataJ).

Centrality:

• CSDA is the most central journal in the sense of the highest (unweighted) degree.

Some further thoughts

All that is just for the fun of it. As mentioned by the authors, citation counts are heavy-tailed, meaning that just a few papers account for much of the citations of a journal while most of the papers account for few citations. As a matter of fact, the total of citations received is mostly driven by a few super-cited papers, and also is the Cited/Citations matrix $\tilde C$ that I use throughout for building the graph. A reason one could put forward about why JRSS B makes it so well is the read papers: for instance, Spiegelhalter et al. (2002), DIC, received alone 11.9% of all JRSS B citations in 2010. Who’d bet the number of citation this new read paper (JRSS A though) will receive?

## [Meta-]Blogging as young researchers

Posted in General, Statistics by Pierre Jacob on 11 December 2014

Hello all,

This is an article intended for the ISBA bulletin, jointly written by us all at Statisfaction, Rasmus Bååth from Publishable Stuff, Boris Hejblum from Research side effects, Thiago G. Martins from tgmstat@wordpress, Ewan Cameron from Another Astrostatistics Blog and Gregory Gandenberger from gandenberger.org

Inspired by established blogs, such as the popular Statistical Modeling, Causal Inference, and Social Science or Xi’an’s Og, each of us began blogging as a way to diarize our learning adventures, to share bits of R code or LaTeX tips, and to advertise our own papers and projects. Along the way we’ve come to a new appreciation of the world of academic blogging: a never-ending international seminar, attended by renowned scientists and anonymous users alike. Here we share our experiences by weighing the pros and cons of blogging from the point of view of young researchers.

## Unfortunate typos in read paper

Posted in General by nicolaschopin on 3 December 2014

Mathieu and I have just realised that the version of our SQMC paper made available on the RSS web site contains several unfortunate typos. In particular, the symbol for “small o” has been replaced by a “big O” by editors. For instance, Theorem 9 should state the QMC beats standard SMC; i.e. the MSE (mean square error) of an SQMC estimator is

$\mathrm{MSE} = o(N^{-1})$

$\mathrm{MSE} = O(N^{-1})$.

Well, that’s a bummer. For now, I recommend anyone to read instead the arxiv version (updated on Monday).

Posted in General by nicolaschopin on 29 October 2014

Another way to generate random particles

Almost 10 months since my latest post? I guess bloggin’ ain’t my thing… In my defense, Mathieu Gerber and I were quite busy revising our SQMC paper. I am happy to announce that it has just been accepted as a read paper in JRSSB. If all goes as planned, we should present the paper at  the RSS ordinary meeting on Dec 10. Everybody is welcome to attend, and submit an oral or written discussion (or both). More details soon, when the event is officially announced on the RSS web-site.

What is SQMC? It is a QMC (Quasi-Monte Carlo) version of particle filtering. For the same CPU cost, it typically generates much more accurate estimators. Interested? consider reading the paper here (more recent version coming soon), checking this video where I present SQMC, or, even better, attending our talk in London!

## momentify R package at BAYSM14

Posted in General, R, Seminar/Conference, Statistics by Julyan Arbel on 20 September 2014

I presented an arxived paper of my postdoc at the big success Young Bayesian Conference in Vienna. The big picture of the talk is simple: there are situations in Bayesian nonparametrics where you don’t know how to sample from the posterior distribution, but you can only compute posterior expectations (so-called marginal methods). So e.g. you cannot provide credible intervals. But sometimes all the moments of the posterior distribution are available as posterior expectations. So morally, you should be able to say more about the posterior distribution than just reporting the posterior mean. To be more specific, we consider a hazard (h) mixture model

$\displaystyle h(t)=\int k(t;y)\mu(dy)$

where $k$ is a kernel, and the mixing distribution $\mu$ is random and discrete (Bayesian nonparametric approach).

We consider the survival function $S$ which is recovered from the hazard rate $h$ by the transform

$\displaystyle S(t)=\exp\Big(-\int_0^t h(s)ds\Big)$

and some possibly censored survival data having survival $S$. Then it turns out that all the posterior moments of the survival curve $S(t)$ evaluated at any time $t$ can be computed.

The nice trick of the paper is to use the representation of a distribution in a [Jacobi polynomial] basis where the coefficients are linear combinations of the moments. So one can sample from [an approximation of] the posterior, and with a posterior sample we can do everything! Including credible intervals.

I’ve wrapped up the few lines of code in an R package called momentify (not on CRAN). With a sequence of moments of a random variable supported on [0,1] as an input, the package does two things:

• evaluates the approximate density
• samples from it

A package example for a mixture of beta and 2 to 7 moments gives that result:

## Beautiful Science: Picturing Data, Inspiring Insight

Posted in General by Pierre Jacob on 19 March 2014

Hey,

There’s a nice exhibition open until May 26th at the British Library in London, entitled Beautiful Science: Picturing Data, Inspiring Insight. Various examples of data visualizations are shown, either historical or very modern, or even made especially for the exhibition. Definitely worth a detour if you happen to be in the area, you can see everything in 15 minutes.

In particular there are nice visualisations of historical climate data, gathered from the logbooks of the English East India company, whose ships were crossing every possible sea in the beginning of the 19th century. The logbooks contain locations and daily weather reports, handwritten by the captains themselves. Turns out the logbooks are kept at the British Library itself and some of them are on display at the exhibition. More info on that project here: oldweather.org.

## Quantitative arguments as hypermedia

Posted in General by jodureau on 16 January 2014

I’m Joseph Dureau, I have been an avid reader of this blog for while now, and I’m very glad Pierre proposed me to share a few things. Until a few months ago, I used to work on Bayesian inference methods for stochastic processes, with applications to epidemiology. Along with fellow colleagues from this past life, I have now taken the startup path, founding Standard Analytics. We’re looking into how web technologies can be used to enhance browsability, transparency and impact of scientific publications. Here’s a start on what we’ve been up to so far.
Let me just make it clear that everything I’m presenting is fully open source, and available here. I hope you’ll find it interesting, and we’re very excited to hear from you! Here it goes..

To date, the Web has developed most rapidly as a medium of documents for people  rather than for data and information that can be processed automatically.
Berners-Lee et al, 2001

Since this sentence was written, twelve years ago, ambitious and collective initiatives have been undertaken to revolutionize what machines can do for us on the web. When I make a purchase online, my email service is able to understand it from the purchase confirmation email, communicate to the online store service, authenticate, obtain information on the delivery, and provide me with a real-time representation of where the item is located. Machines now have the means to process data in a smarter way, and to communicate over it!

However, when it comes to exchanging quantitative arguments, be it in a blog post or in a scientific article, web technology does not bring us much further than what can be done with pen and paper. (more…)

Tagged with: , , ,