09 Heteroscedasticity

What are the possible consequences of heteroscedasticity for OLS estimator?

• It becomes inconsistent

• It becomes biased

• It becomes inefficient

• It becomes not linear

If you wanted to perform White's test for heteroscedasticity to check whether a model with dependent variable wage and independent variable education suffers from heteroscedasticity, which command would you use:

• reg wage education

  white test

• reg wage education

  estat imtest, white

• estat imtest wage education, white

• There is no opportunity to perform this test in Stata.

If you performed White's test in Stata and got the p-value of 0.02. At 5% significance level you would:

• reject the null hypothesis of heteroscedasticity

• reject the null hypothesis of homoscedasticity

• do not reject the null hypothesis of heteroscedasticity

• do not reject the null hypothesis of homoscedasticity

rvpplot command in Stata

• There is no such a command.

• Can be used without prior regression command.

• Gives a histogram as an output.

• Stands for residual-versus-predictor plot

In the presence of heteroscedasticity if White's robust standard errors are used, usually

• Coefficients change, the standard errors are the same.

• Coefficients change, the standard errors become smaller.

• Coefficients do not change, standard errors become greater.

• Coefficients do not change, standard errors become smaller.