A. S. Serafimovich’s Forgotten Novel,City in the Steppe (1912)

La plupart des écrits prérévolutionnaires de A. S. Serafimovich se trouve reléguée à l’ombre par la popularité en Union Soviétique de son roman "classique" Torrent de fer, 1924. Son premier et son plus long roman, Ville dans la Steppe, 1912, est l’une de ses oeuvres les plus achevées; une synthèse tragique et pittoresque de ces forces lointaines et sourdes concrétisées par le régime capitaliste en Russie à la fin du 19ème siècle. Cet article souligne surtout le dépouillement du monde intérieur des personnages principaux du roman – Zakharka et Polynov – que l’écrivain accomplit magistralement avec de l’acuité psychologique et une imagination puissante. Les destins de ces deux hommes opposés s’entrecroisent d’une façon qui ressemble les coups dans une partie d’échecs. Par une antithèse saisissante, Serafimovitch dévoile l’état de ces deux âmes destinées à être détruites parmi les bestialités d’un milieu capitaliste. L’auteur met en jeu ses ressorts psychologiques pour rendre les personnages secondaires et épisodiques également émouvants et mémorables. On trouve dans la caractérisation des faiblesses, des banalités, un certain pessimisme, parfois des exagérations, mais le tout est compensé par les analyses originales et inquiétantes dont l’auteur empreint toutes ses images. Riche en péripéties, sans être confus, le développement des faits dans le roman a une certaine valeur socio-historique.     

Independence in Contingency Tables Using Simplicial Geometry

Frequently, contingency tables are generated in a multinomial sampling. Multinomial probabilities are then organized in a table assigning probabilities to each cell. A probability table can be viewed as an element in the simplex. The Aitchison geometry of the simplex identifies independent probability tables as a linear subspace. An important consequence is that, given a probability table, the nearest independent table is obtained by orthogonal projection onto the independent subspace. The nearest independent table is identified as that obtained by the product of geometric marginals, which do not coincide with the standard marginals, except in the independent case. The original probability table is decomposed into orthogonal tables, the independent and the interaction tables. The underlying model is log-linear, and a procedure to test independence of a contingency table, based on a multinomial simulation, is developed. Its performance is studied on an illustrative example.     

Shiftwork tolerance and 24-h variations in moods

Abstract

The objective of the study was to identify the differences in 24-h variations in moods between three groups of oil-refinery workers: tolerant shiftworkers, intolerant shiftworkers and workers who had never worked in shifts. Each group comprised 29 workers matched by age. The mood measurements were taken during a 24-h period every 2 h starting from 08:00 h. Results were scored for three scales: positive moods, negative moods and fatigue. The two-factor ANOVA revealed the significant main effect of groups for all three moods indicating differences at the average level of 24-h variations in all moods between workers who differed in the degree of tolerance to shiftwork. The significant main effect of time of day was also found for all moods while the reliable interaction between groups and time of day was observed for negative moods only, indicating a different shape of 24-h variations for non-shiftworkers in comparison to shiftworkers.     

A bivariate Sarmanov regression model for count data with generalised Poisson marginals

We present a bivariate regression model for count data that allows for positive as well as negative correlation of the response variables. The covariance structure is based on the Sarmanov distribution and consists of a product of generalised Poisson marginals and a factor that depends on particular functions of the response variables. The closed form of the probability function is derived by means of the moment-generating function. The model is applied to a large real dataset on health care demand. Its performance is compared with alternative models presented in the literature. We find that our model is significantly better than or at least equivalent to the benchmark models. It gives insights into influences on the variance of the response variables.