<![CDATA[Quantile regression as a robust regression method can be used to overcome the impact of unusual cases on regression estimates such as the presence of potential outliers in the data. The purpose of this study was to evaluate the effectiveness of quantile regression in dealing with potential outliers in multiple linear regression compared to ordinary least square (OLS). This study used simulation data in multiple regression model with the number of independent variables (p=3) for different sample sizes (n = 20, 40, 60, 100, 200) and and repeated 1000 times. The effectiveness of the quantile regression method and OLS in estimating β parameters was measured by Mean square error (MSE) and the best model is chosen based on the smallest Akaike Information Criterion (AIC) value. The results showed that in contrast to OLS, quantile regression was able to deal with potential outliers and provided a better estimator with a smaller mean mean square error. Compared to OLS and other quantiles, this study also provides sufficient results that quantile 0.5 provides the best parameter estimate and the best model based on the smallest MSE and AIC values.]]>
Copyrights © 2020
