Modeling of Robust Regression in Breast Tissue Data

In form of analyzing data that are contaminated with outliers, important method is robust regression (RR) and RR provide resistant (stable) results in the presence of outliers. The purpose of this research is to find the relationship between variable and to develop the best model in robust regression by using breast tissue data that were taken from the UCI machine learning repository. Least trimmed square, Sestimation and MM-estimation is used to develop best fit model using Statistical Analysis System (SAS) is applied to model the robust regression. Then, by using S-estimation, LTS and MM-estimation gave the R 2 values were 0.9991, 0.9990 and 0.8185 respectively. Therefore it can be concluded that all three estimations which were Least trimmed square, S-estimation and MM-estimation were comparable.