Flexible amputation models for investigating missing data

• Christian Staerk ; Linda Müller ^[1] ; Andreas Mayr
  1. [1] Koblenz University of Applied Sciences

     Koblenz University of Applied Sciences

     Kreisfreie Stadt Koblenz, Alemania

     
• 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. 438-441
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • We propose exible missing data (\amputation") models based on beta distributed missing probabilities, which are particularly suited for investigating di erent missing mechanisms. In the proposed models the marginal distribution of these probabilities can be directly specified so that deviations from the \missing completely at random" (MCAR) mechanism can be controlled.We illustrate the exibility of the models when applied on a diabetes data set, where the results of a Bayesian multiple imputation method and a complete case analysis are compared with respect to the analysis of the full data set.