Prueba de hipótesis de media constante en geoestadística: una solución con métodos no paramétricos

Geostatistics is a field of spatial statistics. This treats with realizations of stochastics processes spatially indexed. The data are recorded in sites of a region with spatial continuity. In practice for applying geostatistical methods is crucial to evaluate the stationarity assumption. Generally...

Descripción completa

Detalles Bibliográficos
Autor: Cañón Puentes, Mónica Andrea
Tipo de recurso: tesis de maestría
Estado:Versión aceptada para publicación
Fecha de publicación:2019
País:Colombia
Institución:Universidad Nacional de Colombia
Repositorio:Repositorio UN
Idioma:español
OAI Identifier:oai:repositorio.unal.edu.co:unal/76072
Acceso en línea:https://repositorio.unal.edu.co/handle/unal/76072
Access Level:acceso abierto
Palabra clave:310 - Colecciones de estadística general
Monte Carlo
Geoestadística
Métodos no paramétricos
Estacionariedad
Geostatistics
Stationary
Nonparametric
Descripción
Sumario:Geostatistics is a field of spatial statistics. This treats with realizations of stochastics processes spatially indexed. The data are recorded in sites of a region with spatial continuity. In practice for applying geostatistical methods is crucial to evaluate the stationarity assumption. Generally some graphical methods are used to reach this task. In this work is proposed a method for testing the hypothesis of constant mean. The approach considered is based on noon-parametrics methods and Monte Carlo simulation. To verify the performance of the methodology is carried out a simulation study consideering several scenarios of spatial dependence and levels of spatial trend. The test is also applied to a real data set corresponding to measures of physical-chemical variables obtained in an estuary in the north of Colombia. From the analysis with real and simulated data we conclude that the methodology proposed has a good performance. This is particularly usefull in practice when graphical methods are inconclusive.