Abstract
The analysis of Least Squares: LS is often unsuccessful in the case of outliers in the studied phenomena. OLS will lose their properties and then lose the property of Beast Linear Unbiased Estimator (BLUE), because of the Outliers have a bad effect on the phenomenon. To address this problem, new statistical methods have been developed so that they are not easily affected by outliers. These methods are characterized by robustness or (resistance). The Least Trimmed Squares: LTS method was therefore a good alternative to achieving more feasible results and optimization. However, it is possible to assume weights that take into consideration the location of the outliers in the data and determine them accurately. In order to increase the Weighted Least Trimmed Squares: WLTS, the weight of the sample data on the estimation is repeated. In order to perform this research, the need for detection and investigation of the impact of pollution of the Tigris River in Wasit Governorate has been called for by wastewater, particularly Total Dissolved Solids: TDS as dependent variable, and the impact of three covariates Sulfates: SO4, Chloride: Cl and Phosphate: PO4 pollutants. The evaluation was done in in a precise manner and submitted to the competent authorities. In order to achieve this objective, a sample of (91) positions were drawn and checked in the laboratories of Wasit Governorate.