What are the tests for heteroscedasticity?
There are several tests for heteroscedasticity:
1. Breusch-Pagan test: This test checks whether the variance of the residuals is related to the independent variables. If there is a significant relationship, then heteroscedasticity is present.
2. White test: This test is similar to the Breusch-Pagan test but is more general and can account for different types of heteroscedasticity.
3. Goldfeld-Quandt test: This test involves splitting the data into two groups and comparing the variance of the residuals between the two groups. If the variances differ significantly, then heteroscedasticity is present.
4. Park test: This test involves regressing the squared residuals on the explanatory variables and testing whether the coefficients of the explanatory variables are significantly different from zero.
5. Graphical tests: Heteroscedasticity can also be identified by plotting the residuals against the predicted values or the independent variables. If the plot shows a pattern, then heteroscedasticity is present.