Missing values: sparse inverse covariance estimation and?an?extension to sparse regression

We propose an ℓ ₁-regularized likelihood method for estimating the inverse covariance matrix in the high-dimensional multivariate normal model in presence of missing data. Our method is based on the assumption that the data are missing at random (MAR) which entails also the completely missing at random case. The implementation of the method is non-trivial as the observed negative log-likelihood generally is a complicated and non-convex function. We propose an efficient EM algorithm for optimization with provable numerical convergence properties. Furthermore, we extend the methodology to handle missing values in a sparse regression context. We demonstrate both methods on simulated and real data.     

Materializing Strategy in Mundane Tools: the Key to Coupling Global Strategy and Local Strategy Practice?

While much of the literature on strategy and strategy as practice (SaP) focuses on traditional strategic tools, technologies and discursive practices of managers, this paper extends the understanding of strategic change implementation by proposing that mundane material tools, understood as text, translate global strategic discourse in ways that make sense to workers and orchestrate successful global strategy implementation at the local level. Based on a rich case study within one branch of a national bank, this paper demonstrates how a middle manager's materializing practices developed local strategy practice while simultaneously transforming work and producing strategic figures or indicators that satisfied senior management's global strategic change objectives. The contributions of this paper are threefold: (i) it advances the understanding of the multimodality of materiality by identifying the influence of three types of mundane tools produced locally by a middle manager as he performed his sense of the senior managers’ strategic discourse; (ii) it reveals how these three types of physical texts materialized the manager's sense of this discourse, facilitating frontline workers’ engagement and coupling materiality and orality in a coherent way that allowed workers to embody the company's global strategy in their ‘sayings and doings’; and (iii) it highlights the importance of managers’ ability to materialize a strategic discourse.     

Nonparametric Estimation of the Probability of Detection of Flaws in an Industrial Component,from Destructive and Nondestructive Testing Data,Using Approximate Bayesian Computation

We consider the problem of estimating the probability of detection (POD) of flaws in an industrial steel component. Modeled as an increasing function of the flaw height, the POD characterizes the detection process; it is also involved in the estimation of the flaw size distribution, a key input parameter of physical models describing the behavior of the steel component when submitted to extreme thermodynamic loads. Such models are used to assess the resistance of highly reliable systems whose failures are seldom observed in practice. We develop a Bayesian method to estimate the flaw size distribution and the POD function, using flaw height measures from periodic in‐service inspections conducted with an ultrasonic detection device, together with measures from destructive lab experiments. Our approach, based on approximate Bayesian computation (ABC) techniques, is applied to a real data set and compared to maximum likelihood estimation (MLE) and a more classical approach based on Markov Chain Monte Carlo (MCMC) techniques. In particular, we show that the parametric model describing the POD as the cumulative distribution function (cdf) of a log‐normal distribution, though often used in this context, can be invalidated by the data at hand. We propose an alternative nonparametric model, which assumes no predefined shape, and extend the ABC framework to this setting. Experimental results demonstrate the ability of this method to provide a flexible estimation of the POD function and describe its uncertainty accurately.     

Diagnostics of prior-data agreement in applied Bayesian analysis

This article focused on the definition and the study of a binary Bayesian criterion which measures a statistical agreement between a subjective prior and data information. The setting of this work is concrete Bayesian studies. It is an alternative and a complementary tool to the method recently proposed by Evans and Moshonov, [M. Evans and H. Moshonov, Checking for Prior-data conflict, Bayesian Anal. 1 (2006), pp. 893–914]. Both methods try to help the work of the Bayesian analyst, from preliminary to the posterior computation. Our criterion is defined as a ratio of Kullback–Leibler divergences; two of its main features are to make easy the check of a hierarchical prior and be used as a default calibration tool to obtain flat but proper priors in applications. Discrete and continuous distributions exemplify the approach and an industrial case study in reliability, involving the Weibull distribution, is highlighted.     

Vulnerability to violence of girls of the street in Mauritania

This article is concerned with an empirical study of girls of the street in Mauritania. This study is original for three main reasons. First, it investigates Mauritania, a country where there have been very few studies of the phenomenon of children of the street. Secondly, it seems to us that though the academic literature on street children sometimes provides comparisons between girls and boys, very few studies focus specifically on girls of the street. Thirdly, the literature focusing on children of the street generally elucidates the mechanisms that lead some young people to live on the streets. On the contrary to previous literature, our study does not directly focus on mechanisms that lead some young people to live on the streets. Our study focuses on the question of the socio-economic determinants of the risk of violence. In other words, whereas the literature, following the hypothesis of aberrant families, uses violence as an explanatory factor for the phenomenon of children of the street, we attempt here to highlight the explanatory factors for the risk of violence and we treat this risk as an explained variable. A novel finding emerges from this study: there is less risk of violence facing girls of the street if the mother does not work. Conversely, the risk of violence is greater if the mother works and the father stays at home.     

Weather and Smiling Contagion: A Quasi Experiment with the Smiling Sunshine

Studies have shown that pleasant weather conditions (namely, sunshine) favor positive social relationships and improve moods. However, the effect of sunshine on one nonverbal expression that facilitates social relationships (namely, smiling) has never been studied. In a field experiment, men and women walking alone in the street were passed by a male or a female confederate who displayed a smile to the passersby. The contagion effect of smiling was measured. The study was carried out on days that were evaluated as being either sunny or cloudy, but precaution was taken to control the temperature and not to solicit participants when it rained. It was found that the display of a smile results in a smile more often on sunny days. The positive mood induced by the sun may explain such results.