With the increase in hot temperature events in recent years, there is growing interest in measuring the frequency of recurring temperature events. One of the basis for assessing the frequency of recurrence of temperature events is the probability distribution of temperature events, therefore temperature data is needed to produce statistical modeling, especially in determining the best probability distribution. The study intended to estimate the best-fitted probability model for the daily temperature at the Pekanbaru station in Indonesia from 2010 to 2020 using several statistical analyses. Five continuous probability distributions such as Gamma (GM), Slashed Quasi-Gamma (SQG), Three Parameters Quasi Gamma (TPQG), Two Parameter Gamma-Exponential (TPGE), and Modified Log-Logistic (MLL) distributions were fitted for these tasks using the maximum likelihood technique. To determine the model’s fit to the temperature data, several goodness-of-fit tests were applied, including the graphical methos test (density plot) and Numerical criteria method test (AIC, BIC, and -Log Likelihood). The GM and MLL distribution are found to be the best-fitted probability distribution based on goodness-of-fit tests for the daily temperature data at the Pekanbaru station
Read full abstract