The conditional variance function in a heteroscedastic, nonparametric regression model is estimated by linear smoothing of squared residuals. Attention is focused on local polynomial smoothers. Both ...

Mathematics of Computation, Vol. 34, No. 150 (Apr., 1980), pp. 441-463 (23 pages) Constructive proofs and several generalizations of approximation results of J. H. Bramble and S. R. Hilbert are ...

Bernstein polynomial estimation provides a robust nonparametric technique for approximating both density and distribution functions. Based on the properties of Bernstein polynomials, which uniformly ...