Heteroskedasticity

Chapter 8_Review

Course tutor: Ms. Le Thi Ngoc Mai 1

Consequences of Heteroskedasticity

• OLS is still unbiased and consistent under heteroskedastictiy.

• Also, interpretation of R-squared is not changed.

• The usual F tests and t tests are not valid under heteroskedasticity.

• Heteroskedasticity invalidates variance formulas for OLS estimators

• Under heteroskedasticity, OLS is no longer the best linear unbiased estimator

(BLUE).
Heteroskedasticity-robust standard errors

• Example: Hourly wage equation

Heteroskedasticity robust standard errors may be

larger or smaller than their nonrobust counterparts.
The differences are often small in practice.

F statistics are also often not too different.

If there is strong heteroskedasticity, differences may

be larger. To be on the safe side, it is advisable to
always compute robust standard errors.
Breusch-Pagan test for heteroskedasticity

•  Estimate OLS as usual -> obtain residuals -> square the residuals )
• Run regression to obtain

• Calculate F-statistic or LM statistic

• Find p-value using F-statistic or LM statistic: whether or not reject Ho

of homoskedasticity
White test for heteroskedasticity

•  Estimate OLS as usual -> obtain residuals -> square the residuals )
• Run regression to obtain

• Calculate LM statistic

• Find p-value from LM statistic: whether or not reject Ho of

homoskedasticity
Weighted least squares estimation (WLS)

• Heteroskedasticity is known up to a multiplicative constant

The functional form of the
heteroskedasticity is known

Transformed model

• Example: reg sav inc [aweight=1/sqrt(inc)]

Generalized least squares estimation
(GLS)
• 
• Estimate OLS as usual -> obtain residuals -> square the residuals )

• Run regression using squared residuals as the dependent variable and

all independent variables.

• Obtain fitted value of (h)

• Transform the model using weight = 1/h