under heteroskedasticity, is biased down even under homoskedasticity; and it also has variance that depends on the design/structure of the regressors, unlike the s^2(X'X)^{-1}. (See references below). 2. The term ``robust'' is somewhat misleading, and this is one of the very few plac...
The main goal of CTT is to garner information about tests (not test takers). Of particular interest is a test's reliability in a sample, defined as the proportion of variance in the test that is due to variance in the true scores, Rel(Y) = Var(θ)/Var(Y). In IRT, each ...
Comment: On p. 307, you write that robust standard errors “can be smaller than conventional standard errors for two reasons:the small sample bias we have discussed and their higher sampling variance.” A third reason is that heteroskedasticity can make the conventional s.e. upward-biased. ...
Intuitively, when firing the CEO is more costly, the CEO must have lower perceived ability to make firing him worth it. This result does not depend on whether turnover cost c is larger due to a higher firm cost c(firm) or higher effective personal turnover cost c(pers)/κ. B.2. ...
potential outcomes formulation and the structural equations formulation +in that the random variable :math:`Y(t)` is equal to the random variable :math:`g(t, X, W, \epsilon)`, where :math:`X, W, \epsilon` +is drawn from the distribution that generated each sample in the data set. ...