TD_BREUSCH_PAGAN_GODFREY is used to detect the presence of heteroscedasticity in regression analysis. Heteroscedasticity is the situation where the variability of the error term, such as the difference between the observed values and the predicted values, is not constant across all levels of the independent variables.
If heteroscedasticity is present in the data, an ordinary least squares (OLS) estimator can still provide unbiased estimates of the regression coefficients but it can lead to inefficient estimates. This causes the estimated standard errors of the coefficients to be biased and can underestimate the true standard errors. As a result, the hypothesis tests may be unreliable and lead to incorrect conclusions about the significance of the coefficients.
- Small variations so the OLS estimator is unbiased and efficient
- Sample sizes are large so that the OLS estimator is robust to heteroscedasticity
- Estimates are not used for inference but for prediction
TD_BREUSCH_PAGAN_GODFREY involves regressing the squared residuals from the original regression on the independent variables and their squares, and then using the resulting regression to test whether the coefficients on the squared residuals are statistically different from zero.
TD_BREUSCH_PAGAN_GODFREY is used in econometrics and other fields for regression analysis. It is used in conjunction with other diagnostic tests to assess the assumptions and validity of a regression model.
- Use TD_MULTIVAR_REGR to generate an ART containing an ARTFITRESIDUALS layer.
- Use TD_BREUSCH_PAGAN_GODFREY on the produced ART to perform the Breusch Pagan Godfrey test against the residual layer.
- Retrieve the results of TD_BREUSCH_PAGAN_GODREY primary layer.
- Check the NULL_HYPOTHESIS value to determine if there is heteroscedasticity. ACCEPT means that variance is homoscedastic. REJECT means that variance is heteroscedastic.