Compared to some of the methods of estimating parameters distribution Campbell Great extreme value by using the simulation with the practical application of dust storms | ||
Al Kut Journal of Economics Administrative Sciences | ||
Article 1, Volume 0, Issue 7, February 2018, Pages 68-82 | ||
Author | ||
nazar mostafa al saraf and zanib ali abd alhosaen | ||
Abstract | ||
Extreme value distributions received great attention by researchers in various fields in view of the important applications in the field of natural phenomena such as floods, volcanoes, storms, temperatures in the study of certain types of data statistics to life. This research detailed study on the characteristics and advantages of the first type of dividend value extremist, a distribution of Gumbel Distribution Special values Great, and the aim of this research is the comparison between the methods of estimating the parameters and determine the best methods that give the capabilities are good and which is characterized by the least squares line in order to be applied to real data. Research include derive a set of estimation methods including the method of Maximum Likelihood, the usual method of moments, method of linear moments Jackknife method as well as the proposed method which it is the mixed method. In order that the comparison between the estimation methods above were used simulation techniques and by configuring experiments sizes different samples and the values of initial multiple parameters, either the comparison process will be based on statistical measure combines the variance and bias is the mean square error (MSE), and eventually applied The results obtained from the side experimental data on real data represent the frequency of storms in Baghdad, where is the best way Jackknife2 in estimating the parameters of extreme value model. | ||
Statistics Article View: 75 PDF Download: 39 |