Study of the Robust Least Trimmed Square (LTS) Method in Estimating Multiple Linear Regression Parameters for Data Containing Outliers

Abstract

Determining multiple linear regression parameters can use the Ordinary Least Square (OLS) method. The OLS method must meet the assumptions of the Best Linear Un] Estimator (BLUE) to produce a good multiple linear regression equation model which can be seen based on its residual value (remaining square). When estimating using the OLS method if there are outliers in the data set, the OLS method is not effective for producing a good multiple linear regression equation model. The robust least trimmed square (LTS) method is an alternative method that can be used if there are outliers in the data set. The robust least trimmed square method aims to produce an efficient multiple linear regression equation model without eliminating these outliers.