An Alternative Model for Dry-Spell Probability Analysis

Open Access
Publisher Website
Google Scholar
Abstract
The occurrence of sequences of dry days in the wet season was studied for six localities scattered over the State of SāPaulo, Brazil. In order to determine the most suitable probability model for this tropical region, two frequency distribution functions were analyzed: 1) the Markov chain and 2) the truncated negative binomial. In spite of its widespread use, the Markovian model was shown to be less efficient for fitting estimated data with observed data than the truncated negative binomial model. Abstract The occurrence of sequences of dry days in the wet season was studied for six localities scattered over the State of SāPaulo, Brazil. In order to determine the most suitable probability model for this tropical region, two frequency distribution functions were analyzed: 1) the Markov chain and 2) the truncated negative binomial. In spite of its widespread use, the Markovian model was shown to be less efficient for fitting estimated data with observed data than the truncated negative binomial model.

This publication has 0 references indexed in Scilit: