ارسل ملاحظاتك

ارسل ملاحظاتك لنا







Model Selection in Binary Regression Using Nonlocal Prior with a Practical Application

المصدر: مجلة القادسية للعلوم الإدارية والاقتصادية
الناشر: جامعة القادسية - كلية الادارة والاقتصاد
المؤلف الرئيسي: Al-Hamzawi, Rahim (Author)
مؤلفين آخرين: Kadhim, Maytham Saeed (Co-Author)
المجلد/العدد: مج24, ع1
محكمة: نعم
الدولة: العراق
التاريخ الميلادي: 2022
الصفحات: 354 - 361
ISSN: 1816-9171
رقم MD: 1269553
نوع المحتوى: بحوث ومقالات
اللغة: الإنجليزية
قواعد المعلومات: EcoLink
مواضيع:
كلمات المؤلف المفتاحية:
Nonlocal Prior | Variables Selection | Penalized Regression | Shotgun Stochastic Search
رابط المحتوى:
صورة الغلاف QR قانون
حفظ في:
المستخلص: The mater of determine the effective covariates in a binary regression model has got much attention via last year's, Bayesian variables selection approach using nonlocal priors have become more popular. It has a good properties for achieving variables selection and coefficients estimation. In this paper, a new Bayesian approach for simplified shotgun stochastic search with screening has been proposed in binary quantile regression. Our model is depend on the inverse Laplace prior distributions for the binary quantile regression parameters. We compared our proposed model with other methods in same filed via simulation approach and real dataset. Our proposed model has accurate to performing comparison with other methods in estimating coefficient and selecting active variables.

ISSN: 1816-9171