This paper presents a number of consistency results for nonparametric
kernel estimators of density and regression functions and their deriva
tives. These results are particularly useful in semiparametric estimat
ion and testing problems that rely on preliminary nonparametric estima
tors, as in Andrews (1994, Econometrica 62, 43-72). The results allow
for near-epoch dependent, nonidentically distributed random variables,
data-dependent bandwidth sequences, preliminary estimation of paramet
ers (e.g., nonparametric regression based on residuals), and nonparame
tric regression on index functions.