Nonparametric prediction of a Hilbert space valued random variable (Q1081264): Difference between revisions

Asymptotic properties of the prediction of the transformation \(g(\xi_{T+h})\) based on the observations \(\xi_ t\), \(0\leq t\leq T\), of the measurable, strictly stationary Markov process \(\xi_ t\), \(t\in {\mathbb{R}}\), taking values in a measurable space (E,\({\mathcal B})\), is studied where g is a mapping from E into a separable Hilbert space. A statistical nonparametric predictor of \(g(\xi_{T+h})\) generalizes the ''predictogram'' [see \textit{G. Collomb}, Specifying statistical models, Proc. 2nd Franco-Belgian Meet. Stat., Louvain-la-Neuve/Belg. 1981, Lect. Notes Stat. 16, 182-204 (1983; Zbl 0538.62033)]. Some applications of the main results are considered in the paper, too.