We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Space–time modeling of rainfall data.
- Authors
Luis Guillermo Coca Velarde; Hélio S. Migon; Basilio de B. Pereira
- Abstract
Climate variables assume non‐negative values and are often measured as zero. This is just the case when the rainfall level, in the dry season, is measured in a specified place. Then, the stochastic modeling demands the inclusion of a probability mass point at the zero level, and the resulting model is a mixture of a continuous and a Bernoulli distribution.In this article, spatial conditional autoregressive effects dealing with the idea that neighbors present similar responses is considered and the response level is modeled in two stages. The aim is to consider spatial interpolation and prediction of levels in a Bayesian context. Data on weekly rainfall levels measured in different stations at the central region of Brazil, an area with two well‐marked seasons, will be used as an example. A method for comparing models, based on the deviance function, is also implemented. The main conclusion is that the use of space–time models improves the modeling of hydrological and climatological variables, allowing the inclusion of real life considerations such as the influence of other covariates, space dependence and time effects such as seasonality. Copyright © 2004 John Wiley & Sons, Ltd.
- Subjects
RAINFALL; WEATHER; SEASONS; CLIMATOLOGY
- Publication
Environmetrics, 2004, Vol 15, Issue 6, p561
- ISSN
1180-4009
- Publication type
Article
- DOI
10.1002/env.650