On the use of GEV distribution: A case study of temperature in Malaysia

摘要

Extreme maximum temperature of several stations in Malaysia is modeled by fitting monthly maximums to the Generalized Extreme Value (GEV) distribution. The Augmented Dickey Fuller (ADF) and Phillips Perron (PP) tests are used to detect stochastic trends. The Mann-Kendall (MK) test suggests a non-stationary model. Three models are considered for stations with trend. The Likelihood Ratio test shows that Malacca station favours Model 2 (linear in the location parameter) and Kuantan station favours Model 2 and Model 3 (the logarithm of the scale parameter). The Kolmogorov-Smirnov (KS) goodness-of-fit test shows that monthly maximums converge to the GEV distribution. The return period, that is the period where the temperature exceeds the monthly maximum temperature is obtained.