Search In this Thesis
   Search In this Thesis  
العنوان
Some problem applications in the case of skewed distributions :
المؤلف
Badawe, Amal Tolba Mohamed Tolba.
هيئة الاعداد
باحث / أمل طلبه محمد طلبه بدوي
مشرف / فاطمة علي عبد العاطي
مشرف / أشرف فتوح عيطه
مناقش / أشرف فتوح عيطه
الموضوع
Skewness. Skew-Normal Distribution. Skew-Symmetric Distributions. Hidden Truncation Models.
تاريخ النشر
2009.
عدد الصفحات
183 p. :
اللغة
الإنجليزية
الدرجة
ماجستير
التخصص
الإحصاء والاحتمالات
تاريخ الإجازة
01/01/2009
مكان الإجازة
جامعة المنصورة - كلية التجارة - Department of Applied Statistics and Insurance
الفهرس
Only 14 pages are availabe for public view

from 159

from 159

Abstract

The Skewed distribution is a general extension of symmetric distribution with an additional shape parameter. In recent years, significant progress has been made towards the construction of new families of these distributions. A brief summary for the main extensions of skewed distributions is included with a demonstration for the main problems faced in literatures in the case of using skewed distributions in application. In this research, a simulation study for SN distribution, its estimated parameters, expected errors for different sample size, and shape parameter values are included. This research aims to determine the average of Egyptian Gross Domestic Product, GDP, by both fixed and current prices. The data will be fitted with four different types of skewed distributions SN, SGN, SCN, and SNC showing that these are suitable for data more than normal distribution and then choosing the more sufficient for data. Derivatives for the log-likelihood functions of theses distributions with respect to location, scale and shape parameters for each version are derived to get the maximum likelihood estimators numerically using R program. This research concluded that the skewed distributions are better to fit the skewed data such as economics data. The SGN fit GDP data with fixed prices while the SNC fit GDP data with current prices and it will give more accurate results for statistical inference and constructing related models.