In 2013, 15 clusters of mumps were notified in France; 72% (82/114) of the cases had been vaccinated twice with measles-mumps-rubella vaccine. To determine whether the risk of mumps increased with time since the last vaccination, we conducted a case-control study among clusters in universities and military barracks. A confirmed case had an inflammation of a salivary gland plus laboratory confirmation in 2013. A probable case presented with inflammation of a salivary gland in 2013 either lasting for > 2 days or with epidemiological link to a confirmed case. Controls had no mumps symptoms and attended the same university course, student party or military barracks. We collected clinical and vaccination data via web questionnaire and medical records. We calculated adjusted odds ratios (aOR) using logistic regression. 59% (50/85) of cases and 62% (199/321) of controls had been vaccinated twice. The odds of mumps increased for twice-vaccinated individuals by 10% for every year that had passed since the second dose (aOR 1.10; 95% confidence interval (CI): 1.02-1.19; p = 0.02). Mumps immunity waned with increasing time since vaccination. Our findings contributed to the French High Council of Public Health's decision to recommend a third MMR dose during outbreaks for individuals whose second dose dates > 10 years.

A new method for combining several initial estimators of the regression function is introduced. Instead of building a linear or convex optimized combination over a collection of basic estimators r 1 ,..., r M , we use them as a collective indicator of the proximity between the training data and a test observation. This local distance approach is model-free and very fast. More specifically, the resulting nonparametric/nonlinear combined estimator is shown to perform asymptotically at least as well in the L 2 sense as the best combination of the basic estimators in the collective. A companion R package called COBRA (standing for COmBined Regression Alternative) is presented (downloadable on http://cran.r-project.org/web/packages/ COBRA/index.html). Substantial numerical evidence is provided on both synthetic and real data sets to assess the excellent performance and velocity of our method in a large variety of prediction problems.

-Principal curves are nonlinear generalizations of the notion of first principal component. Roughly, a principal curve is a parameterized curve in R d which passes through the "middle" of a data cloud drawn from some unknown probability distribution. Depending on the definition, a principal curve relies on some unknown parameters (number of segments, length, turn. . . ) which have to be properly chosen to recover the shape of the data without interpolating. In the present paper, we consider the principal curve problem from an empirical risk minimization perspective and address the parameter selection issue using the point of view of model selection via penalization. We offer oracle inequalities and implement the proposed approaches to recover the hidden structures in both simulated and real-life data.

a b s t r a c tThis paper deals with the problem of quantization of a random variable X taking values in a separable and reflexive Banach space, and with the related question of clustering independent random observations distributed as X . To this end, we use a quantization scheme with a class of distortion measures called Bregman divergences, and provide conditions ensuring the existence of an optimal quantizer and an empirically optimal quantizer. Rates of convergence are also discussed.