CHAPTER THREE: TWO-VARIABLE REGRESSION MODEL 101
which shows that pS2 is a linear estimator because it is a linear function of Y; actually it is a weighted average of Yi with ki serving as the weights. It can similarly be shown that p91 too is a linear estimator.
Incidentally, note these properties of the weights ki:
1. Since the Xi are assumed to be nonstochastic, the ki are nonstochas-tic too.
4. £ kixi = £ kiXi = 1. These properties can be directly verified from the definition of ki.
Was this article helpful?