Robust Correlation Procedure via Sn Estimator

Abstract

Pearson correlation coefficient is the most widely used statistical technique when measuring a relationship between the bivariate normal distribution when the assumptions are fulfilled. However, this classical correlation coefficient performs poor in the presence of an outlier. Therefore, this study aims to propose a new version of robust correlation coefficient based on robust scale estimator Sn. The performance of the proposed robust correlation coefficient is assessed based on correlation value, average bias and standard error. The performance of the proposed coefficient is compared with the classical correlation together with the existing robust correlation coefficient. Classical correlation coefficient performs well under the condition of perfect data. However, its performance becomes worst when data is contaminated. Under the condition of data contamination, robust correlation coefficient performed better compared to classical correlation.