The Banister model is a widely used model in training (Banister et al., 1975). However, this model does not take into account the population variability. Mixed models arise when repeated measurements are made on subjects (Laird et al., 1982). A model of popular behaviour was constructed, allowing parameters varying from an individual to another, to take into account the heterogeneity between subjects. Results showed that reactions to training were not the same for all subjects. Also, reaction to X2 and X3 varied significantly through the 3 seasons. The proposed model fit significantly training and performance data for all the subjects in the study. Differences between athletes` reactions to training agree with precedent studies (Mujika et al., 1996). In spite of taking into account transformations through seasons, as well as common and individual variability, R2 were not so good than in precedent studies (Mujika et al., 1996). A longer period of study could explain this less quality of fitting values.
© Copyright 2002 Alle Rechte vorbehalten. / Übersetzt mit https://www.DeepL.com/Translator