Adjust of the probabilistic distribution for the multi-year average temperature variable for the department of Boyacá (Colombia) Ajustes de distribuciones probabilísticas para la variable temperatura media multianual para el departamento de Boyacá (Colombia)

Abstract

This paper presents a study on the selection of the best probabilistic distribution for multi-year average temperature variable in the department of Boyacá (Colombia) as the basis for future estimates and projections of the variable under uncertainty arises. Normal, Gamma, Weibull and lognormal distributions were selected to fit the data. To find which best fits the data criteria information based in maximum likelihood, Akaike (Akaike Information Criterion) and Bayesian (Bayesian Information Criteron) were used. The results are shown in both tabular and graphical form, as well as a map of the probabilistic distribution functions most representative in the study area. As a result we obtain that the Weibull distribution is the best fits in general.