Generalized Linear Mixed Models for Monthly Average Temperatures in an Intertropical Region

• Activity Number: 587
• Type: Contributed
• Date/Time: Wednesday, August 3, 2011 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Mercedes Andrade-Bejarano*+ and Edinson Zuluaga and John Jairo Millan and Gabriel Conde-Arango
• Companies: Universidad del Valle and Universidad del Valle and Universidad del Valle and Universidad del Valle
• Keywords: Generalised Linear Models ; Linear Mixed Models ; Isotropic Covariance Function ; ENSO phenomenon

Abstract

Data for this research come from time series of monthly average temperature from 28 sites, collected over the period 1971 to 2002. Due to the geographical location of the study zone (Valle del Cauca, Colombia), monthly average temperature is affected by the altitude and the ENSO phenomenon. Time series for some of the sites show a tendency to increase. Also due to the two dry and wet periods in the study zone, a seasonal behaviour in monthly average temperature is seen. Andrade (2009) modelled monthly average temperature by using linear mixed models, with spatial and temporal correlated errors. However, the errors did not show normal behaviour. The heavy tails of residuals are determined because of the extreme values of temperature in years when the ENSO phenomenon occurs. In this research, we modelled monthly average temperature by using generalized linear mixed models, as an alternative to solve the lack of normality in models fitted by Andrade (2009). We found the Beta distribution as the proper distribution to model the response variable. In the models, years and sites are associated with random effects and altitude, the SOI and the geographical position are fixed effects.