scientific article; zbMATH DE number 4131496

From MaRDI portal
Publication:3033162





[For the entire collection see Zbl 0667.00025.] Let \(X_ n=\theta X_{n-1}+\epsilon_ n\), \(n=1,2,..\). be an AR-1-model with positive or bounded innovations. An estimate \(W_ n=\min_{1\leq k\leq n}X_ k/X_{k-1}\) of the autocorrelation coefficient \(\theta\), based on the extreme value statistic, is proposed. Asymptotic properties are investigated and an asymptotic essentially nonparametric confidence interval for the autocorrelation coefficient is derived.











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3033162)