THE APPLIED OF FUNCTIONAL DATA ANALYSIS TO COMPARISON 100 TIMES SIMULATED MONTHLY RAINFALL USING SOME TWO PARAMETERS DISTRIBUTION

Three probability models of mothly rainfall such as Gamma, Weibull and Log Normal distribution are evaluated in terms their ability to reproduce the mean statistics derived from 100 times simulation monthly Pekanbaru City, Indonesia. One important studies is investigate understand simulate patterns that occur throughout year. To identify pattern, it requires a curve represent observation received during Functional data analysis (FDA) methods capable convert discrete into function can result, try describe hidden rainfall. This study focused on investigating average simulatated by three different quantile functions using FDA. The standard deviation FDA for precipitation obtained. Through these two confidence interval curves presented 95% pointwise intervals. In this study, most actual were around was within interval. simulations gamma log normal distributions found capture historical data, contradictory results shown distribution, cannot be captured outside Based Mean Absolute Error (MAE) value FDA, concluded produce simulated rain better than distribution.