ABSTRACT
LEMMA 5 Assume the columns of X* are linearly independent, F0 (t) has non-vanishing density f 0 (t) and j~(t) is continuous. Then d2(e, F11 , F0) is strictly convex in e.
Lemma 5 indicates that there exists a unique minimizer for the criterion function. It is convenient to define the estimators in terms of estimating equations.
