Modelling between- and within-season trajectories in elite athletic performance data

Athletic performance follows a typical pattern of improvement and decline during a career. This pattern is also often observed within-seasons as athlete aims for their performance to peak at key events such as the Olympic Games or World Championships. A Bayesian hierarchical model is developed to analyse the evolution of athletic sporting performance throughout an athlete's career and separate these effects whilst allowing for confounding factors such as environmental conditions. Our model works in continuous time and estimates both the average performance level of the population, $g(t)$, at age $t$ and how each $i$-th athlete differs from the average $f_i(t)$. We further decompose $f_i(t)$ into changes from season-to-season, termed the between-season performance trajectory, and within-season performance trajectories which are modelled by a constrained Bernstein polynomial. Hence, the specific focus of this project is to identify the differences in performance that exist both between and within-seasons for each athlete. For the implementation of the model an adaptive Metropolis-within-Gibbs algorithm is used. An illustration of algorithm's performance on 100 metres and 200 metres freestyle swimming in both female and male athletes is presented.