Grey system theory is a multidisciplinary scientific approach, which deals with systems that have partially unknown information (small sample and uncertain information). Grey modeling as an important component of such theory gives successful results with limited amount of data. Grey Models are divided into two types; univariate and multivariate grey models. The univariate grey model with one order derivative equation GM (1,1) is the base stone of the theory, it is considered the time series prediction model but it doesn’t take the relative factors in account. The traditional multivariate grey models GM(1,M) takes those factor in account but it has a complex structure and some defects in " modeling mechanism", "parameter estimation "and "m

In this research weights, which are used, are estimated using General Least Square Estimation to estimate simple linear regression parameters when the depended variable, which is used, consists of two classes attributes variable (for Heteroscedastic problem) depending on Sequential Bayesian Approach instead of the Classical approach used before, Bayes approach provides the mechanism of tackling observations one by one in a sequential way, i .e each new observation will add a new piece of information for estimating the parameter of probability estimation of certain phenomenon of Bernoulli trials who research the depended variable in simple regression  linear equation. in addition to the information deduced from the past exper

The search is contain compared among some order selection criteria  (FPE,AIC,SBC,H-Q) for the Model first order  Autoregressive when the White Noise is follow Normal distribution and some of non Gaussian distributions (Log normal, Exponential and Poisson distribution ) by using Simulation  

يناقش هذا البحث مشكلة التعدد الخطي شبه التام في انموذج الانحدار اللاخطي ( انموذج الانحدار اللوجستي المتعدد) ، عندما يكون المتغير المعتمد متغير نوعيا يمثل ثنائي الاستجابة اما ان يساوي واحد لحدوث استجابة او صفر لعدم حدوث استجابة ، من خلال استعمال مقدرات المركبات الرئيسية التكرارية(IPCE)  التي تعتمد على الاوزان الاعتيادية والاوزان البيزية الشرطية .

اذ تم تطبيق مقدرات هذا ا

In this paper, we will provide a proposed method to estimate missing values for the Explanatory variables for Non-Parametric Multiple Regression Model and compare it with the Imputation Arithmetic mean Method, The basis of the idea of this method was based on how to employ the causal relationship between the variables in finding an efficient estimate of the missing value, we rely on the use of the Kernel estimate by Nadaraya – Watson Estimator , and on Least Squared Cross Validation (LSCV) to estimate the Bandwidth, and we use the simulation study to compare between the two methods.