Welitom Ttatom Pereira Da Silva,Versides Sebastião Moraes e Silva,LUIZ AIRTON GOMES,Mariano Martínez Espinosa, Universidade de Brasília, Engenharia Civil, wttatom@terra.com.br
Abstract
The main aim of this study was to determine a statistical model capable of projecting a per capita water quota for the city of Cuiabá, Mato Grosso State, Brazil. The sample was composed of the 115 recognized neighborhoods of Cuiabá, separated by month and referring to the period between 2003 and 2006. The methodology used was based on the statistical model (analysis of regression and verification of the models). As a result we obtained a model stratified based on socioeconomic status with reasonable projection capability of per capita share to the Cuiabá city. A combination of grouping techniques using the models of Burtless and Hausman (1978) or artificial intelligence techniques could be an alternative for obtaining better models.
Keywords: water management, statistical modeling, water resources