1) If I suspect that heteroskedasticity is present in the following model .ie., we don't believe that the variance of women's earnings would be the same as the variance in men's earnings, what am I suspecting to be true (i.e. define heteroskedasticity)? What OLS assumption is violated if this were true? What are the repercussions regarding my analysis if I do not correct for this problem?
Earningsi = B0 + B1 × ui
Ordinary least squares (OLS) is a method estimating an unknown parameter of a variable in a regression model. The value it produces is called beta. It assumes that the variance of the error term is constant. Heteroskedasticity exists when the variance of the ...
How Heteroskedasticity affects the outcome of significance tests are determined.