|
Activity Number:
|
423
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, August 2, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #320816
|
View Presentation
|
|
Title:
|
Spatial-Temporal Pareto Modeling of Extreme Value Data
|
|
Author(s):
|
Gabriel Huerta* and Luis Enrique Nieto Barajas
|
|
Companies:
|
University of New Mexico and Instituto Tecnologico Autonomo de Mexico
|
|
Keywords:
|
Autoregressive model ;
latent variables ;
Pareto distribution ;
Pollution data ;
Mexico City ;
stationarity
|
|
Abstract:
|
In this work we introduce a spatio-temporal process with Pareto marginal distributions for extreme value observations. The dependence in space and time of the process is introduced through the use of latent variables in a hierarchical fashion. For certain specifications the process becomes strictly stationary in space and time. We review the construction of the process and highlight some of its properties. We follow a Bayesian approach to estimate model parameters and produce predictions. The performance of the model is illustrated with a pollution data set of monthly maxima ozone concentrations in the metropolitan area of Mexico City. Results show that our model is superior to two other alternative models.
|
Authors who are presenting talks have a * after their name.
