Serviços Personalizados
Journal
Artigo
Espanhol (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por emailIndicadores
-
Citado por SciELO -
Acessos
Links relacionados
-
Citado por Google -
Similares em
SciELO -
Similares em Google
Compartilhar
Permalink
Revista Integración
versão impressa ISSN 0120-419X
Integración - UIS vol.35 no.1 Bucaramanga jan./jun. 2017
https://doi.org/10.18273/revint.v35n1-2017006
Articulo Original
Geoestadística aplicada a series de tiempo autorregresivas: un estudio de simulación
Geostatistics applied to autorregresive time series: A simulation study
1 Departamento de Estadística, Universidad Nacional de Colombia, Bogotá, Colombia.
2 Departamento de Matemáticas, Escuela Colombiana de Ingeniería Julio Garavito, Bogotá, Colombia.
3 Turkish Petroleum Company, Bogotá, Colombia.
La geoestadística puede usarse como método de predicción de datos faltantes en series temporales. El procedimiento se basa en el estudio de la estructura de autocorrelación temporal de la serie de tiempo por medio de la función de variograma, que es estimada por mínimos cuadrados (geoestadística clásica) o por máxima verosimilitud (geoestadística basada en modelo), y en usar posteriormente Kriging para hacer predicción de los valores faltantes. En este trabajo se comparan a través de simulación las dos aproximaciones (geoestadística clásica y basada en modelo) en el contexto de series temporales autorregresivas.
Palabras clave: Autocorrelación; Datos faltantes; Kriging; Predicción; Semivariograma; Series de tiempo; Variograma
Geostatistics can be used as a method for predicting missing data in time series. The procedure is based on estimating the temporal autocorrelation structure by means of the semivariance function, by least squares (classical geostatistics) or maximum likelihood (model-based geostatistics), and posteriorly using Kriging for doing prediction of missing data in the time series. In this work we compare classical and model-based geoestatistics in the context of time series using simulated autorregresive time series.
Keywords: Autocorrelation; Kriging; Prediction; Missing values; Semivariogram; Time series; Variogram
Referencias
1. Box G., Jenkins G. and Reinsel G., Time Series Analysis: Forecasting and Control, John Wiley & Sons, New York, 2008. [ Links ]
2. Cressie N., Statistical for Spatial Data, John Wiley & Sons, New York, 1993. [ Links ]
3. De Iaco S., Palma M. and Posa D., “Geostatistics and the Role of Variogram in Time Series Analysis: A Critical Review”, en Statistical Methods for Spatial Planning and Monitoring ( ed. Montrone S. and Perchinunno P.), Springer-Verlag, Milan, 2013. [ Links ]
4. Diggle P. and Ribeiro P., Model-based Geostatistics, Springer-Verlag, New York, 2007. [ Links ]
5. Dowdall M., Bjorn L., Gerland S. and Rudjord A., “Geostatistical analysis as applied to two environmental radiometric time series”, Environmental Monitoring and Assessment 83 (2003), No. 1, 1-16. [ Links ]
6. Ghahraman B. and Ahmadi F., “Application of geostatistics in time series: Mashhad annual rainfall”, Iran Watershed Management Science and Engineering 1 (2007), No. 1, 7-15. [ Links ]
7. Haslett J., “On the sample variogram and sample autocovariance for non-stationary time series”, Journal Royal Statistical Society Series D 46 (1997), No. 4, 475-484. [ Links ]
8. Hollander T. and Wolfe D., Nonparametric Statistical Methods, John Wiley & Sons, New York, 1999. [ Links ]
9. Ljung G., “A note on the estimation of missing values in time series”, Comm. Statist. Simulation Comput. 18 (1989), No. 2, 459-465. [ Links ]
10. Ma Ch., “The use of the variogram in construction of stationary time series models”, J. Appl. Probab. 41 (2004), No. 4, 1093-1103. [ Links ]
11. Margalef R., Limnología, Ediciones Omega, Barcelona, 1983. [ Links ]
12. Peña D. and Tiao G., “A note on likelihood estimation of missing values in time series”, Amer. Statist. 45 (1991), No. 3, 212-213. [ Links ]
13. R Core Team, R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna, 2013. [ Links ]
14. Ribeiro P. and Diggle P., “geoR: a package for geostatistical analysis”, R-NEWS 1 (2001) , No. 2, 15-18. [ Links ]
15. Rouhani S. and Myers D., “Problems in space time Kriging of geohidrological data”, Mathematical Geology 5 (1990), No. 22, 611-623. [ Links ]
16. Schabenberger O. and Gotway C., Statistical Methods for Spatial Data Analysis, Chapman & Hall, Boca Ratón, 2005. [ Links ]
Recibido: Abril de 2016; Aprobado: Mayo de 2017
Este es un artículo publicado en acceso abierto bajo una licencia Creative Commons
