المصدر: | المجلة العلمية للاقتصاد والتجارة |
---|---|
الناشر: | جامعة عين شمس - كلية التجارة |
المؤلف الرئيسي: | El Gohary, Mervat (AUTH.) |
المجلد/العدد: | ع1 |
محكمة: | نعم |
الدولة: |
مصر |
التاريخ الميلادي: |
2007
|
الصفحات: | 21 - 29 |
ISSN: |
2636-2562 |
رقم MD: | 664509 |
نوع المحتوى: | بحوث ومقالات |
اللغة: | الإنجليزية |
قواعد المعلومات: | EcoLink |
مواضيع: | |
رابط المحتوى: |
المستخلص: |
The method of least squares for estimating the parameters of the linear regression model is a widely used technique. It is efficient under certain conditions. However, in the presence of heavy tailed errors or outliers, the least squares efficiency is reduced. Robust regression techniques have been proposed to overcome this problem. Compound estimation is one of these techniques. It is based on one step multi-stage generalized M estimators. This paper proposes two compound estimators based on a variation of the initial estimator and the leverage measure. Their performance under a variety of error distributions is examined via Mont Carlo simulations. |
---|---|
ISSN: |
2636-2562 |