On the selection of number of knots in linear regression splines with free-knots

• Gioia Di Credico ^[1] ; Francesco Pauli ^[1] ; Nicola Torelli
  1. [1] University of Trieste

     University of Trieste

     Trieste, Italia

     
• Localización: Proceedings of the 35th International Workshop on Statistical Modelling : July 20-24, 2020 Bilbao, Basque Country, Spain / Itziar Irigoien Garbizu (ed. lit.), Dae-Jin Lee (ed. lit.), Joaquín Martínez Minaya (ed. lit.), María Xosé Rodríguez Álvarez (ed. lit.), 2020, ISBN 978-84-1319-267-3, págs. 334-337
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • Linear regression splines are useful tools to describe departures from linearity in several real applications. Location of knots can be seen as change points in the relationship between the variables. In a Bayesian context, we analyze the variation of the Stochastic Search Variable Selection approach previously proposed in Di Credico et al. (2018), focusing on the impact of the hyperparameters choice on the estimation of the correct number of knots.