| Autor: |
Rana, Sohel, Muhammad, Adamu Adamu, Elgohari, Hanaa |
| Zdroj: |
International Journal of Agricultural & Statistical Sciences; Dec2022, Vol. 18 Issue 2, p697-703, 7p |
| Abstrakt: |
Goldfeld-Quandt test appears to be a powerful test of heteroscedasticity when the assumption upon which the test was built holds. The test procedure requires ordering observations of one explanatory variable such that the arrangement of observations from other variables in the regression model follows. However, for multiple explanatory variables, identifying this suitable variable to be used in the ordering becomes a problem when there is no prior knowledge of which variable causes the heteroscedasticity problem. This study has proposed a way of identifying this variable prior to conducting the Goldfeld-Quandt test. The performance of the proposed procedure is evaluated by the simulation study and real data set. The results show that the proposed procedure successfully identifies the explanatory variables which need to be used in the Goldfeld-Quandt test procedure. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
