keywords: Linear estimation; Least squares; Error; Power of estimation
The linear estimation and the least squares estimation are known estimation procedures. This study tends to investigate the robustness of these procedures. To determine the robustness of linear and least squares estimation procedures, the correlation of the two procedures are determined. Thereafter, the power of estimation(P_ca ) is applied to determine the robustness or accuracy of the estimation procedures using the correlation values and the sample size of the data set. Robustness of the estimation procedures is determined if P_ca>0.5 otherwise if P_ca<0.5 implies that the variation between the original data and the estimated data set is relatively large. The results revealed that the correlation and the value of P_ca demonstrated that the two estimation procedures are perfect. However, for this study, the average P_ca=0.735 which means that estimated values are robust this concludes that the estimated values of the data set is robust for the linear and least square estimation procedures.