Let m(n)(x) and M(n)(x) be a partitioning estimate and the kernel esti
mate, respectively, of a regression function m(x) = E(Y/X = m) for the
i.i.d. sample (X(1),Y-1),...,(X(n), Y-n). Under the condition E\Y\(p)
< infinity, where p > I, and some conditions on the partition and the
kernel function, the strong L(1)-consistency is proved.