R. Meléndez, R. Giraldo
The Sign test is a non-parametric alternative used for testing hypothesis in one sample and two paired samples problems, when the classical tests based on the normality assumption are not appropriated. In the literature of FDA there are some approaches for testing hypothesis when we have one sample and two paired samples coming from Gaussian processes. When the sample is not Gaussian and n is small, it is necessary to use other approaches. In this paper we give an alternative in this sense. In particular we propose the Sign test for functional data as a method for testing hypothesis if we have one sample and two paired samples of functional data. Based on a simulation study, we show that the proposed test has a good performance. We illustrate the methodology by applying it to a meteorological data set.
Palabras clave / Keywords: Functional data, hypothesis test, preorder, Sign test
Programado
Sesión V03 Análisis de Datos Funcionales
1 de junio de 2018 16:00
Sala 2
