مقارنة بين طريقي الامكان الأعظم والبيزية في تقدير أنموذج الانحدار الذاتي المكاني مع تطبيق عملي على مرض السرطان في العراق

Abstract EN

Recently, spatial data analysis has attracted the attention of statistical researchers, especially because neglecting the spatial dimension in the analysis affects the results and leads to the loss of important information.

therefore, spatial regression models have been resorted to, through which it is possible to study the extent to which the dependent variable is affected by the explanatory variables in light of the presence of spatial dependence for the studied vocabulary.

the bayes method and the MLE method were used to estimate the spatial autoregressive model (SAR) in the presence of the modified spatial weights matrix that was built based on rook juxtaposition criterion.

cancer patient data on the number of cancer patients in each province of Iraq have been relied upon as an approved variable, but the illustrative variables were the age rate, the rate of tumor size of cancer patients and the number of areas contaminated with uranium in the province, and through the results of the mean absolute percentage error criteria (MAPE) that used to compare the methods of estimation it was found that the Maximum Likelihood method (MLE) is better than bayes method to estimate the spatial auto regressive model in the absence of a modified spatial weight matrix built up based on the Rook juxtaposition standard.