Asymptotic normality of sums of Hilbert space valued random elements (Q2041371)

The main object of study in the current work is the random series \(Z_n:=\sum_{k\in\mathbb{Z}}u_{n,k}X_k, \ n\in\mathbb{N}\), where \((X_k)_{k\in\mathbb{N}}\) may be either a sequence of i.i.d random variables with values in a separable Hilbert space \(\mathcal{H}\) or it may constitute a linear proces with i.i.d. innovations. The coefficients \((u_{n,k})_{k\in\mathbb{Z}}\) are linear bounded operators in \(\mathcal{L}(\mathcal{H})\). If the random variables \((X_k)\) are centered with finite second moments, and the sequence of weights \((u_{n,k})\) satisfies a series of assumptions that ensure the almost sure convergence of the series \((Z_n)_{n\in\mathbb{N}}\), then in the main result Theorem 2.2, the author identifies the limit of \((Z_n)\). The main result is then extended to the case when \((X_k)_{k\in\mathbb{Z}}\) is a linear process with innovations in Section 3. Convergence of discounted sums process and nearly non-stationary models are investigated as applications of the main result. Series of examples are provided as well.