The Equivalence of Two Proportions Revisited

The classic conditional test for checking that the difference between two independent proportions is not null may not be appropriate in many circumstances. Dunnett & Gent (1977) showed that in clinical trials, in studies of drugs, etc, the aim is to prove the practical equality (equivalence) of both proportions. On other occasions the aim may be the opposite: i.e. to prove that the two proportions are substantially different (biologically significant). Both cases are usually solved by two one-sided tests (TOST test). In this article, this procedure is shown to be conservative and two true two-sided tests for each case are proposed.     

Analysis of an addictions treatment system

A treatment system consisting of an assessment/referral service, a detox, a mission, a hospital recovery unit, and three recovery homes was examined using a retrospective client-tracking approach over 1 year. There were 4,187 admissions to the system accounted for by 1,522 individuals; system users tended to be males, unmarried, unemployed, and in their 30s and 40s. The best predictor of readmission to the system after treatment was prior system use. Use patterns and referral data indicated that perceived gatekeepers (i.e., detoxication and assessment services) were not consistently coordinating the system and that more socially stable clients appeared to be more likely to be involved with non-addiction specific services (i.e., medical, general counselling).     

Testing fit for the grouped exponential distribution

Grouped data can often arise due to the lack of resolution of the measurement instruments; they also arise when data are deliberately rounded to a certain accuracy and are presented, say, in the form of a histogram. The author uses statistics of the Cramér‐von Mises type to test for the exponential distribution when data are grouped.     

Dependence Estimation for High‐frequency Sampled Multivariate CARMA Models

The paper considers high‐frequency sampled multivariate continuous‐time autoregressive moving average (MCARMA) models and derives the asymptotic behaviour of the sample autocovariance function to a normal random matrix. Moreover, we obtain the asymptotic behaviour of the cross‐covariances between different components of the model. We will see that the limit distribution of the sample autocovariance function has a similar structure in the continuous‐time and in the discrete‐time model. As a special case, we consider a CARMA (one‐dimensional MCARMA) process. For a CARMA process, we prove Bartlett's formula for the sample autocorrelation function. Bartlett's formula has the same form in both models; only the sums in the discrete‐time model are exchanged by integrals in the continuous‐time model. Finally, we present limit results for multivariate MA processes as well, which are not known in this generality in the multivariate setting yet.     

China: Towards the introduction of dependency/long‐term care insurance

The Chinese social security system has been the subject of numerous publications, which have made policy developments more accessible to researchers and administrators from all countries. However, the steps introduced in response to growing demands for intervention by the authorities in favour of dependent persons have remained poorly documented in the international literature. The purpose of this article is to take stock of pilot experiments in this field since the beginning of the 13th Five‐Year Plan (2016–2020) with regard to their policy objective, operating mode and financing modalities.     

L'évaluation par le client comme méthode de surveillance: les enjeux de la réputation numérique du travailleur à la lumière du Règlement général sur la protection des données de l'Union européenne

L'évaluation du travailleur par le client ou l'employeur ainsi que la publication des avis qui en résultent constituent une méthode de contrôle et de surveillance qui pose de multiples problèmes au regard du droit européen. Sous couvert d'une approche client, elle soumet le travailleur à une surveillance de tous les instants, qui va bien au‐delà de ce qui prévaut dans un cadre de travail classique. L'auteur examine les enjeux de la question à la lumière du Règlement général sur la protection des données de l'Union européenne (UE), qui fixe des limites très précises à la collecte et la publication d'informations concernant les travailleurs.     

Négociation et performance: une analyse exhaustive et précise de l'effet de différents systèmes de négociation sur la productivité des entreprises

S'appuyant sur des données représentatives au niveau des entreprises, disponibles pour l'ensemble de l'UE, les auteurs analysent la relation entre les processus et structures institutionnelles de négociation collective et l'évolution de la productivité du travail. Ils montrent que la nature du système de négociation a une influence marquée: certaines caractéristiques sont associées à une baisse de performance (notamment la présence d'un système de négociation sectorielle non coordonné), d'autres à une amélioration (systèmes sectoriels coordonnés). Plus que le caractère individuel ou collectif de la négociation, c'est donc bien la nature des processus et structures dans lesquels elle s'inscrit qui importe en la matière.     

Estimands in practice: Bridging the gap between study objectives and statistical analysis

The International Council for Harmonisation (ICH) guideline E9 Statistical Principles for Clinical Trials (1) was issued in 1998. In October 2014, an addendum to ICH E9 was proposed on statistical principles relating to estimands and sensitivity analyses. The final version of the addendum to ICH E9 (R1) (2) was issued in November 2019. This virtual edition of Pharmaceutical Statistics takes a closer look at some of the progress that has been made since 2018 when implementing the estimand framework within clinical research. The articles discussed in this virtual issue are not new, but a compilation from previous issues. This specific article will act as a refresher for those not familiar with the topic and discuss the ABCs of estimands and their proposed deployment for improving the quality of clinical research. An overview of the more recent Pharmaceutical Statistics articles on estimands will be provided, signifying areas where progress have been made. The articles should be considered as contributions to the ongoing discussions rather than the final word. Finally, a personal perspective on the estimand success story and remaining challenges with proposed solutions will be discussed.     

Progrès économiques et avancées sociales au sein des chaînes d'approvisionnement mondiales: plusieurs exemples dans les pays du cône Sud

Les auteurs s'interrogent sur la nature et la diffusion des progrès induits par les chaînes d'approvisionnement mondiales (CAM), en s'appuyant sur les conclusions de neuf études de cas portant sur des filières implantées dans des pays du sud de l'Amérique latine. Des progrès économiques, associés à certaines avancées en matière sociale, sont observés parmi les entreprises têtes de file, mais les fournisseurs en amont n'en tirent pas toujours profit. Après avoir isolé trois types d'évolutions possibles au sein des CAM (développements «antagoniste», «tronqué» et «intégré»), les auteurs formulent des recommandations en ce qui concerne l'action des pouvoirs publics, des entreprises et des acteurs sociaux.     

Non-exclusive group contests: An experimental analysis

We experimentally study a non-exclusive group contest in which contestants actively participate in multiple groups simultaneously. We compare the results of this contest to those of an exclusive group contest in which each contestant belongs to a single group. In contrast to theoretical predictions, we find that the non-exclusive group contest generates less aggregate effort than the equivalent exclusive group contest. We hypothesize that groups in the non-exclusive group contest are less responsive to their rival group’s effort than those in the exclusive group contest. Likewise, on the individual level, players in the non-exclusive group contest are more likely to free-ride on their group members’ contributions. Our data indicate that non-free-riders in the non-exclusive group contest are more likely, over time, to allocate their effort toward a single group. This finding is consistent with previous findings that players facing a complex strategy space tend to focus on specific winning combinations. Moreover, given that players are affected by their group members’ contributions, they tend to exert their effort primarily toward a single group. Taken together, our findings suggest that a non-exclusive group contest may evolve, over time, into an exclusive group contest.