From intersection local time to the Rosenblatt process (Q895915)

\textit{M. S. Taqqu} [Z. Wahrscheinlichkeitstheor. Verw. Geb. 31, 287--302 (1975; Zbl 0303.60033)] introduced the Rosenblatt process with parameter \(H\in(1/2,1)\) as a limit in distribution of partial sum processes of strongly dependent random variables. It is the simplest non-Gaussian Hermite process and constitutes a counterpart of fractional Brownian motion, which is the most important long-range dependent Gaussian process. In this paper, the authors present a construction of the Rosenblatt process by means of an appropriate particle system, intersection local time and white noise analysis. They also introduce a dependence exponent of the Rosenblatt process which turns out to be \(2-2H\) and hence is the same as for fractional Brownian motion with Hurst parameter \(H\).