Which pattern indicates heteroscedasticity?

Which pattern indicates heteroscedasticity?

• A. Residual variance increases with fitted values
• B. Residual variance constant across fitted values
• C. Residuals evenly distributed around zero
• D. Residuals are normally distributed

Answer: (A)

Heteroscedasticity shows up when the variability of the residuals changes with the level of the fitted values. In a residuals vs. fitted-values view, you’d see the spread of residuals get wider as the fitted values increase. That increasing variance is exactly what pattern described here indicates. If the residual spread stays roughly the same across all fitted values, that’s homoscedasticity. If residuals cluster in a way that they sit around zero with no obvious pattern, there isn’t a visible heteroscedastic pattern. Normality of residuals relates to the shape of their distribution, not to whether their variance changes with the fitted value.

Heteroscedasticity shows up when the variability of the residuals changes with the level of the fitted values. In a residuals vs. fitted-values view, you’d see the spread of residuals get wider as the fitted values increase. That increasing variance is exactly what pattern described here indicates.

If the residual spread stays roughly the same across all fitted values, that’s homoscedasticity. If residuals cluster in a way that they sit around zero with no obvious pattern, there isn’t a visible heteroscedastic pattern. Normality of residuals relates to the shape of their distribution, not to whether their variance changes with the fitted value.