The conditional central limit theorem in Hilbert spaces. (Q2574610)

The conditional central limit theorem of the authors [Ann.\ Probab.\ 30, 1044--1081 (2002; Zbl 1015.60016)] is studied for dependent random variables. For a measure preserving transformation \(T\) on a Hilbert space \(\mathbf H\) the sequence \(X_i = X_0\circ T^i\) and the process \(S_n=X_1 + \dots + X_n\) are defined. In the first part of the paper equivalent conditions for the conditional CLT for \(S_n\) and for the functional CLT for \(W_n(t) = S_{[nt]} + (nt-[nt])X_{[nt]+1}\) are found. In the second part the results are applied to weakly dependent sequences and to \(\mathbf H\)-valued linear processes.