Modeling swimming marks through Blocks and POT methods

• Author(s): Gomes, Dulce ; Teles, Júlia ; Canto e Castro, Luísa
• Date: 2011
• Persistent ID: http://hdl.handle.net/10174/4309
• Origin: Repositório Científico da Universidade de Évora
• Subject(s): Extreme value models; Generalized Pareto models; Blocks method; Peaks over threshold method
• Document Type: Article
• Language: English
• Editor(s): Conesa, David; Forte, Anabel; López-Quílez, Antonio

Description

The swimming marks in the 100 m men's freestyle long course are modelled using extreme value theory. Using the statistical package R, extreme value and generalized Pareto models were adjusted in order to estimate the left endpoint of these models. The left endpoint can be interpreted as the best mark that can ever be reached, admitting that swimming pool conditions, athlete's equipment and training methods remain the same.