Modeling the distribution of rainfall amount / Isnewati Ab Malek ... [et al.]

The amount of rainfall from seven selected stations in Negeri Sembilan used to find the best fit model employing four continuous distributions: Exponential, Gamma, Weibull, and Normal. This study analyzed distributions of monthly rainfall amounts for ten years from the year 2010 until 2019. Parameters for each distribution were estimated using the maximum likelihood method. Also, the model selection technique based on Akaike Information Criterion (AIC), and Bayesian Information Criterion (BIC) have been used in examining the best fitting distribution among the four distributions. The Anderson-Darling goodness-of-fit test was applied to all the distributions to determine if the data set follows the specified distribution or not. The test indicates that the Weibull and Normal distribution can be used to model the rainfall amount in Negeri Sembilan. Hence, the Weibull distribution is the best model in describing the rainfall amounts since it has the best fit among all criteria of both AIC and BIC as the distribution has the lowest value compared to other distributions.