Convergence rates for kernel regression in infinite-dimensional spaces (Q2304253)

The authors have discussed a nonparametric regression problem, where the covariate \(X\) is a random element in a complete separable metric space, and the response \(Y\) lies in some arbitrary measure space. In this paper, the optimum convergence rate for a wide class of kernel regression estimates when the covariate as well as the response may be infinitedimensional has been derived, and the problem of optimum convergence rates of nonparametric regression estimates has been investigated by the authors.