I’m just back from the Society for Research on Educational Effectiveness meetings, where I presented work on small-sample corrections for cluster-robust variance estimators in two-stage least squares models, which I’ve implemented in the clubSandwich R package. Here’s my presentation. So I had “clubSandwich” estimators on the brain when a colleague asked me about whether the methods were implemented in SAS.
The short answer is “no.”
The moderately longer answer is “not unless we can find funding to pay someone who knows how to program properly in SAS.
In settings with independent observations, sample size is one way to quickly characterize the precision of an estimate. But what if your estimate is based on weighted data, where each observation doesn’t necessarily contribute to equally to the estimate? Here, one useful way to gauge the precision of an estimate is the effective sample size or ESS. Suppose that we have (N) independent observations (Y_1,…,Y_N) drawn from a population with standard deviation (\sigma), and that observation (i) receives weight (w_i).