Wavelet-based estimation of multivariate regression functions in Besov spaces

Let (Y,X) = {Y-i,X-i} be real-valued jointly stationary processes and let r ho be a Borel measurable function on the real line. Let g(x) = E[rho(Y-1)\X -1 = x] be a d-dimensional regression function. For regression functions in the Besov space B-s,B-p,B-q we estimate g using orthonormal wavelet bases. Uniform rates of almost sure convergence over compact subsets of R-d are e stablished for strongly mixing processes.