A New Extension of Generalized Extreme Value Distribution: Extreme Value Analysis and Return Level Estimation of the Rainfall Data

This paper presents an extension of the generalized extreme value (GEV) distribution, based on T-X family distributions: Gompertz-generated distributions that make existing distribution more flexible called Gompertz-general (Go-GEV) distribution. Some properties proposed are introduced, and a new is applied to actual data, namely rainfall in Lopburi Province, by comparing model with traditional GEV estimating return levels Province. Results showed Go-GEV was alternative for values fitted data described maximum better than The probability density functions had various shapes including left-skewed, right-skewed close symmetric. Estimation Province indicated Buachum Station should be monitored because it higher precipitation at 2, 5, 10 15 years. HIGHLIGHTS analysis complex but ignoring observation just unacceptable not best practice. One widely used tools such situations “Extreme Value Theory” Gompertz-generalized using deffinition developed its function diverse shape, as both symmetrical applying GRAPHICAL ABSTRACT