A1162
Title: Mixed effects quantile autoregressive modeling for point-referenced daily maximum temperatures in Aragon, Spain
Authors: Jorge Castillo-Mateo - University of Zaragoza (Spain) [presenting]
Ana C Cebrian - University of Zaragoza (Spain)
Jesus Asin - University of Zaragoza (Spain)
Alan Gelfand - Duke University (United States)
Abstract: Different spatial patterns of climate change are analyzed across quantiles associated with point-referenced daily maximum temperatures in Aragon, northeastern Spain. For that purpose, regression through asymmetric Laplace errors is considered in the context of a very flexible mixed effects autoregressive model, introducing two temporal scales and four spatial processes. Moreover, while the autoregressive model yields conditional quantiles, it is demonstrated how to extract marginal quantiles from the conditional quantiles with the asymmetric Laplace specification. Marginal quantiles enjoy direct interpretation as well as the benefit of spatial interpolation, i.e., they do not require the previous day's temperature.
