PENERAPAN METODE NEWEY WEST DALAM MENGOREKSI STANDARD ERROR KETIKA TERJADI HETEROSKEDASTISITAS DAN AUTOKORELASI PADA ANALISIS REGRESI

Abstract

Ordinary Least Squares (OLS) used to estimate the parameters in the regression analysis. If one of the assumptions is not fulfilled, the results of the OLS are no longer best, linear, and unbiased properties. The aim of this research was to find out the application of Newey West method to correct standard error when heteroscedasticity and autocorrelation occurred, and to compare the results of OLS with Newey West method on secondary and simulation data. OLS can still be used to estimate the regression parameter when heteroscedasticity and autocorrelation occurred. However, it will cause bias on standard error of parameter. A method which can correct the standard error of parameters to be unbiased parameter is Newey West method. The secondary data about Passenger Car Milage and data simulated contain heteroscedasticity and autocorrelation. The analysis showed that the Newey West method were known is able to correct standard error when heteroscedasticity and autocorrelation occurred on both of data. It was obtained that Newey west method with and changes the value of the bias standard error of OLS to be unbiased.