Heteroscedasticity
- Heteroscedasticity, is a statistical term used in regression analysis.
- It describes a situation where the variability of the errors (residuals) in a regression model is not constant across all levels of the independent variable(s).
1) Increasing Heteroscedasticity: In this model, the difference of the residuals increases as the values of the independent variables increase. This manner that as we move along the predictor variables, the spread of the residuals becomes extensive.
2)Decreasing Heteroscedasticity: In contradiction to increasing heteroscedasticity, this model involves the variance of the residuals decreasing as the values of the independent variable increase. The expansion of residuals channels as you move along the predictor variable.
3)U-shaped Heteroscedasticity: U-shaped heteroscedasticity occurs when the spread of residuals forms a particular U shape as you move along the independent variables. The dissimilarity of residuals is not constant and emerges to be heteroscedastic in a systematic manner.
The Breusch pagan test
- It is a statistical test applied in regression analysis to verify for the existence of heteroscedasticity in a regression model.
- The test is to determine whether the residuals is constant across all levels of the independent variables.
Null Hypothesis: If there is no heteroscedasticity the difference of outcomes remains constant .
Alternative Hypothesis: If there is heteroscedasticity the difference of outcomes is not constant and may vary across the tosses.
- If P value is greater than the given significance level then it is not subjected to reject the null hypothesis.
- On the other hand, If P value is less than the given significance level then it is subjected to reject the null hypothesis.