توضيف تقنية المويجي في تقدير نموذج الإنحدار اللامعلمي المعلم (WGAM)‎ : دراسة مقارنة مع المحاكاة و التطبيق

Abstract EN

In this paper, the problem of the unknown probabilistic distribution of data and employing the method of estimating the non-parametric generalized additive models (GAM) based on smoothing splines as smoothers, and dealing with this case by using an iterative approach.

the wavelet shrinkage technique was used and relied upon to estimate the wavelet generalized additive models (WGAM) regression model, which was proposed to be used as a data smoothers through the use of some wavelets as filters in the calculation of the discrete wavelet transformation, and some statistical criteria were used to compare the estimation methods.

this done through the use of simulation and analysis of real data as well, the efficiency of the proposed methods was tested on data collected from Ibn Sina teaching hospital, on cases of short stature, and wavelet shrinkage smoothers gave the best results compared to the usual GAM method and the wavelet helped on smoothing data, by obtaining the most efficient results.