Bayesian sampling without tears

Following a question on Stack Overflow trying to replicate a figure from the paper written by Alan Gelfand and Adrian Smith (1990) for The American Statistician, Bayesian sampling without tears, which precedes their historical MCMC papers, I looked at the R code produced by the OP and could not spot an issue as to why their simulation did not fit the posterior produced in the paper. Which proposes acceptance-rejection and sampling-importance-resampling as two solutions to approximately simulate from the posterior. The later being illustrated by simulations from the prior being weighted by the likelihood… The illustration is made of 3 observations from the sum of two Binomials with different success probabilities, θ¹ and θ². With a Uniform prior on both.

for (i in 1:N)
  for (k in 1:3){
    llh<-0
    for (j in max(0,n2[k]-y[k]):min(y[k],n1[k]))
      llh<-llh+choose(n1[k],j)*choose(n2[k],y[k]-j)*
         theta[i,1]^j*(1-theta[i,1])^(n1[k]-j)*theta[i,2]^(y[k]-j)*
         (1-theta[i,2])^(n2[k]-y[k]+j)
    l[i]=l[i]*llh}

To double-check, I also wrote a Gibbs version:

theta=matrix(runif(2),nrow=T,ncol=2)
x1=rep(NA,3)
for(t in 1:(T-1)){
   for(j in 1:3){
     a<-max(0,n2[j]-y[j]):min(y[j],n1[j])
     x1[j]=sample(a,1,
        prob=choose(n1[j],a)*choose(n2[j],y[j]-a)*
           theta[t,1]^a*(1-theta[t,1])^(n1[j]-a)*
           theta[t,2]^(y[j]-a)*(1-theta[t,2])^(n2[j]-y[j]+a)
           )}
      theta[t+1,1]=rbeta(1,sum(x1)+1,sum(n1)-sum(x1)+1)
      theta[t+1,2]=rbeta(1,sum(y)-sum(x1)+1,sum(n2)-sum(y)+sum(x1)+1)}

which did not show any difference with the above. Nor with the likelihood surface.

COMPUTO, the journal for reproducible statistical research

Computo (latin for calculate, compute, reckon) is a new journal launched by the French statistical society (SFDS) to promote reproducible research in statistics and machine learning by publishing papers with reproducible contributions. Towards this goal, Computo goes beyond classical static publications by including technical advances in literate programming and scientific reporting. The reproducibility of numerical results is a necessary condition for publication in Computo. In particular, submissions must include all necessary data (e.g. via zenodo repositories) and code. For contributions featuring the implementation of methods/algorithms, the quality of the provided code is assessed during the review process. Meaning accepting contributions in the form of notebooks (e.g. Rmarkdown, or Jupyter). This is a 100% free and open-access journal, thanks to the sponsoring of the SFDS. Once a manuscript is accepted, its reviews will be made available on the Computo website. Reviewers can choose to remain anonymous or not. (Towards an even broader reach, we are now considering a partnership with the PCI, following an earlier attempt I did not pursue till its completion…) Computo’s logo has been designed by Loïc Schwaller. And represents the letters of Computo in bytes. Submissions are now open!

reproducibility check [Nature]

While reading the Nature article Swarm Learning, by Warnat-Herresthal et [many] al., which goes beyond federated learning by removing the need for a central coordinator, [if resorting to naïve averaging of the neural network parameters] I came across this reporting summary on the statistics checks made by the authors. With a specific box on Bayesian analysis and MCMC implementation!

news from PCI

on anonymisation

An article in the New York Times covering a recent publication in Nature Communications on the ability to identify 99.98% of Americans from almost any dataset with fifteen covariates. And mentioning the French approach of INSEE, more precisely CASD (a branch of GENES, as ENSAE and CREST to which I am affiliated), where my friend Antoine worked for a few years, and whose approach is to vet researchers who want access to non-anonymised data, by creating local working environments on the CASD machines  so that data does not leave the site. The approach is to provide the researcher with a dedicated interface, which “enables access remotely to a secure infrastructure where confidential data is safe from harm”. It further delivers reproducibility certificates for publications, a point apparently missed by the New York Times which advances the lack of reproducibility as a drawback of the method. It also mentions the possibility of doing cryptographic data analysis, again missing the finer details with a lame objection.

“Our paper shows how the likelihood of a specific individual to have been correctly re-identified can be estimated with high accuracy even when the anonymized dataset is heavily incomplete.”

The Nature paper is actually about the probability for an individual to be uniquely identified from the given dataset, which somewhat different from the NYT headlines. Using a copula for the distribution of the covariates. And assessing the model with a mean square error evaluation when what matters are false positives and false negatives. Note that the model need be trained for each new dataset, which reduces the appeal of the claim, especially when considering that individuals tagged as uniquely identified about 6% are not. The statistic of 99.98% posted in the NYT is actually a count on a specific dataset,  the 5% Public Use Microdata Sample files, and Massachusetts residents, and not a general statistic [which would not make much sense!, as I can easily imagine 15 useless covariates] or prediction from the authors’ model. And a wee bit anticlimactic.