Extremes of alpha-ARCH models

MaRDI QIDQ4518948zbMATH OpenFDO

Authors C. Y. Robert

Publication date 11 May 2001

zbMATH Keywords

alpha-ARCHARCHextremesstationarytailtime series

Mathematics Subject Classification ID

Statistical methods; economic indices and measures (91B82) Economic time series analysis (91B84)

Review text

The process which is considered is defined by the autoregressive scheme \[ X_t=\sqrt{a+b(X_{t-1}^2)^\alpha}\varepsilon_t,\quad t\in \mathbb{Z}, \] where \(a,b >0\), \(\alpha\in (0,1)\), \(\{\varepsilon_t,t\in \mathbb{Z}\}\) is i.i.d with standard gaussian distribution. The pattern of the tails for \(\alpha > 1/2\) is studied. It is shown that the extremal index exists and the law of the asymptotic normalized maximum is given. An application for the Alcatel stocks is given. For the entire collection see [Zbl 0943.00038].

Recommendations

• scientific article; zbMATH DE number 5866253
• Extremes of Stochastic Volatility Models
• Extremes of stochastic volatility models
• Extreme Value Theory for GARCH Processes
• A Class of Nonlinear Arch Models
• Long-range dependence effects and ARCH modeling
• scientific article; zbMATH DE number 774844

Cited in

(3)

• Properties of some statistics for AR-ARCH model with application to technical analysis
• Extremal behaviour of solutions to a stochastic difference equation with applications to ARCH processes
• Joint exceedances of the ARCH process

This page was built for publication: Extremes of alpha-ARCH models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4518948)