Percentiles curves based on multivariate conditional transformation models.Application to diabetes

• Autores: Óscar Lado Baleato, Carmen María Cadarso Suárez , Thomas Kneib , Francisco Gude Sampedro
• 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. 137-141
• Idioma: inglés
• Enlaces
  • Texto Completo Libro
• Resumen

  • Multivariate Conditional Transformation Models (MCTMs) were recently proposed as a new multivariate regression technique. MCTMs characterize jointly the covariates e ects on the marginal distributions of the responses and their correlations. Flexibility, in both the responses and covariates e ects are achieved using Bernstein polynomial basis. Based on MCTMs, in this paper percentile curves are constructed for each response. Simulation studies indicated the good performance of these estimated condtional percentiles. Finally, MCTMs percentile curves were obtained for three diabetes markers (fasting plasma glucose, glycated hemoglobin and fructosamine) condtionally on age.