Archive ouverte UNIGE | last documents for author 'Stephane Heritier'https://archive-ouverte.unige.ch/Latest objects deposited in the Archive ouverte UNIGE for author 'Stephane Heritier'engDiscussion of “the power of monitoring: how to make the most of a contaminated multivariate sample” by andrea cerioli, marco riani, anthony c. atkinson and aldo corbellinihttps://archive-ouverte.unige.ch/unige:100294https://archive-ouverte.unige.ch/unige:100294This paper discusses the contribution of Cerioli et al. (Stat Methods Appl, 2018), where robust monitoring based on high breakdown point estimators is proposed for multivariate data. The results follow years of development in robust diagnostic techniques. We discuss the issues of extending data monitoring to other models with complex structure, e.g. factor analysis, mixed linear models for which S- and MM-estimators exist or deviating data cells. We emphasise the importance of robust testing that is often overlooked despite robust tests being readily available once S- and MM-estimators have been defined. We mention open questions like out-of-sample inference or big data issues that would benefit from monitoring.Wed, 13 Dec 2017 16:50:25 +0100Practical applications of bounded-influence testshttps://archive-ouverte.unige.ch/unige:96034https://archive-ouverte.unige.ch/unige:96034This chapter discusses the practical applications of bounded-influence tests. The robust versions of classical likelihood ratio, Wald or score tests, are now available in a general setting. They are more reliable than their classical counterparts—that is, they are not influenced by small deviations from the underlying model and can also be used as useful diagnostic tools to identify influential or outlying data points. The chapter illustrates their performance to show that they can be easily implemented in different practical situations. They can be used to robustly choose a model when the hypotheses are non-nested. That is when the model under the null hypothesis cannot be obtained as a particular or limiting case of the model under the alternative hypothesis. The approach followed is the approach based on the influence function. It is mainly concerned with the local robustness properties of tests. A parametric model is considered to study the effects of departures from the model on the testing procedures. The chapter also discusses robust testing in generalized linear models (or GLIM) and emphasizes the use of robust tests in logistic regression. As typical examples, the Food–Stamp data analyzed by Stefanski, Carroll, and Ruppert and the data introduced by Cormier, Magnan, and Morard in auditing is used.Mon, 07 Aug 2017 13:52:30 +0200The role of adaptive trial designs in drug developmenthttps://archive-ouverte.unige.ch/unige:95062https://archive-ouverte.unige.ch/unige:95062Clinical development of new drugs is a long and costly process. There is a need to find solutions which can improve and shorten this process. By introducing flexibility in to the design of clinical trials, adaptive design contributes to this improvement and allows to reach drug development decisions in a quicker way. Areas covered: We review the main methodological approaches to adaptive trial design, introducing key statistical concepts. For each phase of the clinical development, different uses and implementations of adaptive trial (AD) design are presented and examples of recent clinical trials are given. The guidance documents issued by the US and European regulatory authorities are also presented. Expert commentary: Despite inevitable challenges, prospects of this rapidly evolving approach to drug development are important. Controlling the risk of type 1 error and the potential operational risks which may be associated with adaptive trial strategy is paramount in late phase studies. However, with new methodological work, these risks are now well controlled and adaptive trial design will certainly shape the future of drug development.Mon, 26 Jun 2017 11:25:41 +0200Saddlepoint tests for accurate and robust inference on overdispersed count datahttps://archive-ouverte.unige.ch/unige:89099https://archive-ouverte.unige.ch/unige:89099Inference on regression coefficients when the response variable consists of overdispersed counts is traditionally based on Wald, score and likelihood ratio tests. As the accuracy of the p-values of such tests becomes questionable in small samples, three recently developed tests are adapted to the negative binomial regression model. The non-trivial computational aspects involved in their implementation, some of which remained obscure in the literature until now, are detailed for general M-estimators. Under regularity conditions, these tests feature a relative error property with respect to the asymptotic chi-squared distribution, thus yielding highly accurate p-values even in small samples. Extensive simulations show how these new tests outperform the traditional ones in terms of actual level with comparable power. Moreover, inference based on robust (bounded influence) versions of these tests remains reliable when the sample does not entirely conform to the model assumptions. The use of these procedures is illustrated with data coming from a recent randomized controlled trial, with a sample size of 52 observations. An R package implementing all tests is readily available.Fri, 18 Nov 2016 13:59:00 +0100Contributions to overdispersed count data modeling: robustness, small samples and other extensionshttps://archive-ouverte.unige.ch/unige:45953https://archive-ouverte.unige.ch/unige:45953This PhD thesis contributes to the modeling of overdispered count data in three ways. First, we extend two approaches for building robust M-estimators of the regression parameters in the class of generalized linear models to the negative binomial (NB) distribution. Second, we adapt recently developed tests, such as the so-called saddlepoint test, to the framework of overdispersed count data and give a detailed account of their computation and implementation. Through extensive simulations we compare them to traditional tests in order to assess their effective level under the null hypothesis and their power under alternatives, and this for models based on a full NB likelihood or on moment restrictions only. Finally, we enlarge our scope to the class of mixed compound Poisson models, where the overdispersion is due to the variance of unobserved mixing variables. We propose a semi-parametric framework where the mixing distribution is left unspecified and estimated by point-masses.Wed, 28 Jan 2015 11:24:05 +0100Robust inference in the negative binomial regression model with an application to falls datahttps://archive-ouverte.unige.ch/unige:40977https://archive-ouverte.unige.ch/unige:40977A popular way to model overdispersed count data, such as the number of falls reported during intervention studies, is by means of the negative binomial (NB) distribution. Classical estimating methods are well-known to be sensitive to model misspecifications, taking the form of patients falling much more than expected in such intervention studies where the NB regression model is used. We extend in this paper two approaches for building robust $M$-estimators of the regression parameters in the class of generalized linear models to the NB distribution. The first approach achieves robustness in the response by applying a bounded function on the Pearson residuals arising in the maximum likelihood estimating equations, while the second approach achieves robustness by bounding the unscaled deviance components. For both approaches, we explore different choices for the bounding functions. Through a unified notation, we show how close these approaches may actually be as long as the bounding functions are chosen and tuned appropriately, and provide the asymptotic distributions of the resulting estimators. Moreover, we introduce a robust weighted maximum likelihood estimator for the overdispersion parameter, specific to the NB distribution. Simulations under various settings show that redescending bounding functions yield estimates with smaller biases under contamination while keeping high efficiency at the assumed model, and this for both approaches. We present an application to a recent randomized controlled trial measuring the effectiveness of an exercise program at reducing the number of falls among people suffering from Parkinson's disease to illustrate the diagnostic use of such robust procedures and their need for reliable inference.Sun, 19 Oct 2014 12:26:06 +0200Robust Bounded-Influence Tests in General Parametric Modelshttps://archive-ouverte.unige.ch/unige:23217https://archive-ouverte.unige.ch/unige:23217We introduce robust tests for testing hypotheses in a general parametric model. These are robust versions of the Wald, scores, and likelihood ratio tests and are based on general M estimators. Their asymptotic properties and influence functions are derived. It is shown that the stability of the level is obtained by bounding the self-standardized sensitivity of the corresponding M estimator. Furthermore, optimally bounded-influence tests are derived for the Wald- and scores-type tests. Applications to real and simulated data sets are given to illustrate the tests' performance.Sat, 06 Oct 2012 21:33:17 +0200Robust Binary Regression with Continuous Outcomeshttps://archive-ouverte.unige.ch/unige:22947https://archive-ouverte.unige.ch/unige:22947Les auteurs proposent un estimateur à fonction d'influence bornée pour la régression binaire à variable dépendante continue, qui constitue une alternative à la régression logistique lorsque le chercheur s'intéresse à la proportion d'individus pour lesquels la variable expliquée est supérieure ou inférieure à un certain seuil. Les auteurs montrent à la fois théoriquement et empiriquement que dans ce contexte, l'estimation à vraisemblance maximale est sensible à une mauvaise spécification du modèle. Ils montrent que leur estimateur robuste est plus stable et presqu'aussi efficace que le maximum de vraisemblance si les hypothèses sont satisfaites. De plus, l'inférence à laquelle il conduit est plus sûre. Les auteurs comparent la performance des différents estimateurs par voie de simulation et présentent une analyse de l'hypertension sur des données en provenance de Harlem.Tue, 18 Sep 2012 11:24:20 +0200Robust Methods in Biostatisticshttps://archive-ouverte.unige.ch/unige:22633https://archive-ouverte.unige.ch/unige:22633Robust statistics is an extension of classical statistics that specifically takes into account the concept that the underlying models used to describe data are only approximate. Its basic philosophy is to produce statistical procedures which are stable when the data do not exactly match the postulated models as it is the case for example with outliers. Robust Methods in Biostatistics proposes robust alternatives to common methods used in statistics in general and in biostatistics in particular and illustrates their use on many biomedical datasets. The methods introduced include robust estimation, testing, model selection, model check and diagnostics. They are developed for the following general classes of models: Linear regression. Generalized linear models. Linear mixed models. Marginal longitudinal data models. Cox survival analysis model. The methods are introduced both at a theoretical and applied level within the framework of each general class of models, with a particular emphasis put on practical data analysis. This book is of particular use for research students, applied statisticians and practitioners in the health field interested in more stable statistical techniques. An accompanying website provides R code for computing all of the methods described, as well as for analyzing all the datasets used in the book.Wed, 29 Aug 2012 07:55:36 +0200