Three-step N–R algorithm for the maximum-likelihood estimation of the general extreme values distribution parameters

2008 ◽  
Vol 39 (5) ◽  
pp. 384-394 ◽  
Author(s):  
Tefaruk Haktanir
Author(s):  
Valentin Raileanu ◽  

The article briefly describes the history and fields of application of the theory of extreme values, including climatology. The data format, the Generalized Extreme Value (GEV) probability distributions with Bock Maxima, the Generalized Pareto (GP) distributions with Point of Threshold (POT) and the analysis methods are presented. Estimating the distribution parameters is done using the Maximum Likelihood Estimation (MLE) method. Free R software installation, the minimum set of required commands and the GUI in2extRemes graphical package are described. As an example, the results of the GEV analysis of a simulated data set in in2extRemes are presented.


2017 ◽  
Vol 4 (2) ◽  
pp. 8-14
Author(s):  
J. A. Labban ◽  
H. H. Depheal

"This paper some of different methods to estimate the parameters of the 2-Paramaters Weibull distribution such as (Maximum likelihood Estimation, Moments, Least Squares, Term Omission). Mean square error will be considered to compare methods fits in case to select the best one. There by simulation will be implemented to generate different random sample of the 2-parameters Weibull distribution, those contain (n=10, 50, 100, 200) iteration each 1000 times."


Author(s):  
Valentin Raileanu ◽  

The article briefly describes the fields of application of the theory of extreme values, including climatology. The data format and the analysis methods of the annual maxima and minima temperatures in Chisinau are presented. Free R software and the GUI in2extRemes graphical package are used. Estimating the parameters of the Generalized Extreme Value distribution and return levels vs return periods is done using the Maximum Likelihood Estimation (MLE) method.


Author(s):  
Haitham Yousof ◽  
S. Jahanshahi ◽  
Vikas Kumar Sharma

In this paper, we investigate a new model based on Burr X and Fréchet distribution forextreme values and derive some of its properties. Maximum likelihood estimation alongwith asymptotic confidence intervals is considered for estimating the parameters of thedistribution. We demonstrate empirically the flexibility of the distribution in modelingvarious types of real data. Furthermore, we also provide Bayes estimators and highestposterior density intervals of the parameters of the distribution using Markov ChainMonte Carlo (MCMC) methods.


Sign in / Sign up

Export Citation Format

Share Document