Uji Park Dan Uji Breusch Pagan Godfrey Dalam Pendeteksian Heteroskedastisitas Pada Analisis Regresi

Abstract

Homoskedastisitas is one of the conditions are fulfilled classical assumptions in the regression analysis, if not met this means homoskedastisitas error variance is not constant and is said to occur heteroscedasticity problem. Test Park and Pagan Godfrey Breusch test is a statistical test to detect whether there is a problem of heteroscedasticity in the regression equation. The problem is how to test the results of detection heteroskedastisitas Park and Breusch Pagan Godfrey test, which is more effective test.Based on the results of research and discussion can be concluded that the detection of the three cases of data acquired two pieces of data in the test with both test detected heteroskedasticity problems, while one case is detected by the test Breusch heteroskedastisitas Pagan Godfrey Park but the test was not detected. Values mean square error (MSE) test Breusch Pagan Godfrey smaller than the test Park so it can be said Pagan Godfrey Breusch method used more effectively. Thus, in detecting problems hetereoskedastisitas should use Breusch Pagan Godfrey test because they have better accuracy than tests Park.