B. Pulido Bravo, A. M. Franco-Pereira, R. E. Lillo
When working with functional data a problem arises when the aim is to order functions. There are different concepts available in the literature to tackle this problem. The statistical depth provides a criterion to order them from center to outwards, while the epigraph and hypograph indices give an ordination from top to bottom or vice versa.
This work proposes new definitions of these indices based on areas between curves. This new approach better isolates the outlying curves and can be considered in several data analysis problems, such as outlier detection or clustering. Finally, the good performance of these indices is presented through synthetic and real datasets with a special focus on environmental datasets.
Palabras clave: Epigraph, hypograph, depth, functional data, outlier detection, clustering
Programado
Análisis de Datos Funcionales II
12 de junio de 2025 11:30
Sala VIP Jaume Morera i Galícia