Asymptotic theory for curve-crossing analysis (Q886113): Difference between revisions

The authors consider the following object: \[ C_n(\mu)=\sum_{i=1}^n{\mathbf 1}_{(X_{i-1}-\mu_{i-1})(X_i-\mu_i)\leq 0}, \] a curve-crossing process, where \((X_i)_{i\in \mathbb Z}\) is a random process such that \(X_n=g(\varepsilon_n,\varepsilon_{n-1},\dots),\) (\(g\) is a measurable function and \(\varepsilon_i, i\in \mathbb Z,\) are i.id. random variables) and \(\mu=(\mu_i)_{i\in \mathbb Z}\) is a sequence of real numbers. Central limit theorems are obtained and the asymptotic distributions are shown to be either multiple Wiener-Ito integrals or integrals with respect to stable Lévy processes.