Data observations often contain anomalous cases, also known as outliers. These may strongly influence the results and form a fundamental problem in statistical data analysis. Generally speaking, an outlier can lead to model misspecification, incorrect analysis results and can make all estimation procedures meaningless. Thus, in this paper, we supplied a comprehensive method of an alternative for detection of outliers based on robust regression namely weighted MM-estimator. The main purpose of robust regression weighted  MM-estimation is to provide resistant analysis in the presence of outliers in the set of data. An efficient algorithm has been proposed to solve a regression problem. A simulation study was performed to prove that the alternative approach gives better results to identify the outliers by using SAS computer software. This proposed method can be applied to detect an outlier and high leverage points.

outlier, SAS, MM-estimation, robust regression.