Facultade de Fisioterapia

2012/06/05_Marta Sestelo (University of Vigo) and Luis Meira-Machado (University of Minho)

June 5th 2012

Faculty of Economic and Business Sciences | Salón de Grados

Testing critical points of regression curves. An application to the management of aquatic living resources

2012/06/05 - 10:15 h | Marta Sestelo, University of Vigo


This paper was focused on regression models incorporating the so-called factorby-curve interaction, where the effect of a continuous covariate on response varies across groups defined by levels of a categorical variable. This study sought to compare regression curves and their derivatives that may vary across groups defined by different experimental conditions. The goals of this paper were a) to provide a global test to detect significant features of regression curves through the study of their derivatives, and b) to draw inferences about critical points (such as maxima or change points) linked to the derivative curves. The regression curves were estimated using local polynomial kernel smoothers. Such nonparametric regression models allow for a more flexible fit of real data than do the parametric regression techniques usually used. Similarly, they make it possible for the derivatives of the regression curve to be calculated. Bootstrap methods were used to draw inferences from the derivative curves, and binning techniques were applied to speed up computation in the estimation and testing processes. A simulation study was conducted to assess the validity of these bootstrap-based tests. This methodology was applied to study the relative growth of barnacles, in particular, in the estimation of the minimum size of capture of this species. This is a joint work with J. Roca-Pardiñas.

Conditional Transition Probabilities in a non-Markov Illness-death Model

2012/06/05 - 11:00 h | Luis Meira-Machado, University of Minho (Portugal)


One important goal in multi-state modeling is the estimation of transition probabilities. In longitudinal medical studies these quantities are particularly of interest since they allow for long-term predictions of the process. In recent years significant contributions have been made regarding this topic. However, most of the approaches assume independent censoring and do not account for the influence of covariates. This paper introduces feasible estimation methods for the transition probabilities in an illness-death model conditionally on current or past covariate measures. These approaches are evaluated through a simulation study, comparing two different estimators. The proposed methods are illustrated using real data. This is a joint work with S. Datta and J. de Uña-Álvarez.