استعمال بعض طرائق التقدير اللامعلمي لأنموذج البيانات المزدوجة في حالة وجود مشكلة عدم التجانس لتقدير الموازنة العامة للدول العربية للمدة (2002-2014)‎

Abstract EN

In this research has addressed to the study of some estimators nonparametric to model random effects nonparametric double data (Nonparametric Panel Data Models With Random Effects) in case of a problem of heterogeneity, as there is a serious problem is in the selection of display primaries parameter, and therefore we need to find a A quick way to determine the accurate boot by eating standard intersection of legitimacy to the parameter (legitimate transit), and was eating both destined hard topical nonparametric using expansion Taylor (Local Constant Nonparametric Estimator by a Taylor expansion) (LCNE), and the estimated linear topical nonparametric using expansion Taylor ( Local Linear Nonparametric Estimator by a Taylor expansion) (LLNE), and an estimated two-stage estimation method (Nonparametric Two - Step Estimation Method in panel data) (Two - Step), by applying these methods on realistic data indicate the general budget for the Arab States for the period (2014-2002) , and the results proved destined least squares linear positional weighted to model and superiority over the estimated others preference by receiving a lower value to standards (RAMSE), (AMAPE) .

Keyword Panel data: LCNE ,LLNE , Two-step , Bandwidth , RAMSE , AMAPE Random effects.