ارسل ملاحظاتك

ارسل ملاحظاتك لنا







الرئيسية > USING SMALL SAMPLE PERFORMANCE...>الوصف

USING SMALL SAMPLE PERFORMANCE TO TEST THE MAXIMUM LIKELIHOOD ESTIMATES

المصدر: المجلة العلمية للاقتصاد والتجارة
الناشر: جامعة عين شمس - كلية التجارة
المؤلف الرئيسي: Elatraby, Amr I. A. (AUTH.)
المجلد/العدد: ع1
محكمة: نعم
الدولة: مصر
التاريخ الميلادي: 2008
الصفحات: 35 - 52
ISSN: 2636-2562
رقم MD: 664516
نوع المحتوى: بحوث ومقالات
قواعد المعلومات: EcoLink
مواضيع:
نموذج الأداء | احتمالات الحد الاقصى | الاختبارات
رابط المحتوى:
PDF (صورة)     
صورة الغلاف QR قانون

عدد مرات التحميل

3

حفظ في:
المستخلص: The interval estimate for the intraclass correlation based on the maximum likelihood estimate (MLE) has been studied under the beta-binomial distribution (BBD). An adjusted interval estimate based on the MLE is developed to improve the performance of the interval estimate. The small sample performance of this interval estimate for the intraclass correlation is examined by Monte Carlo simulations under a variety of situations. Both balanced sampling, where all clusters have the same size, and unbalanced sampling, where the cluster size varies, are considered. The simulation study shows the interval estimate performs poorly in many situations, especially for those situations where the underlying mean response probability (<0.4) is small. The adjusted interval estimate based on the MLE presents good results in almost all of the situations, except for the cases when both the number of clusters (=20) and the underlying intraclass correlation (=0.1) are small. An example is also included to demonstrate the application of the interval estimate in Business.
ISSN: 2636-2562

عناصر مشابهة


© 2024 المنظومة. جميع الحقوق محفوظة.