Log pearson III distribution and gumbel distribution model for rainfall data in pekanbaru

Abstract

Rainfall is very important for human life. Both high and low rainfall can cause some problems, ranging from flood to drought. We can see that flood can be found in several big cities, and one of them is in Pekanbaru. Therefore, rainfall data analysis is needed to know the model and the pattern of rainfall distribution. In turn, the result can be used as a consideration for the development of the city or the formulation of policy in Pekanbaru’s governance. In this study, Log Pearson III distribution and Gumbel distribution model were used to analyze data of rainfall in Pekanbaru from January 2011 to April 2017. The purpose of this study was to determine the best rainfall model among Log Pearson III and Gumbel Distribution. To estimate parameters from both models, we used Maximum Likelihood and Newton Rephson method. Based on the use of the AIC and BIC test, this study results showed that Log Pearson III is better than Gumbel distribution in modeling the rainfall data of Pekanbaru.