ABSTRACT
Assume that we observe the values of a response variable Y at n predetermined values of an independent variable t. The resulting bivariate observations ( h, Yl ), ... , ( tn, Yn) follow the non parametric regression model
Yi = J.L(ti) + ei, i = 1, ... , n, (3.1)
where e = (e1 , . .. ,en)T is a vector of zero mean, uncorrelated, random errors having common variance u2 , J.L is an unknown regression function that we wish to estimate and 0 :::; t1 :::; · · · :::; tn :::; 1. In this chapter we encounter our first variety of non parametric regression estimator known as a series estimator.
