Asymptotic behaviour for the extreme values of a linear regression model

MaRDI QIDQ1769296zbMATH OpenFDO

Authors Aliou Diop, Dominique Guegan

Publication date 21 March 2005

Published in African Diaspora Journal of Mathematics (Search for Journal in Brave)

zbMATH Keywords

extreme value theorypoint processPoisson random measureregression model

Mathematics Subject Classification ID

Extreme value theory; extremal stochastic processes (60G70) Statistics of extreme values; tail inference (62G32) Linear regression; mixed models (62J05) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10)

Review text

Consider the linear regression model defined by the following relation: \[ Y_t=aX_t+b+\zeta_t, \] where \(a>0\), \((\zeta_t)\) is a white noise process and \((X_t)\) is a sequence of independent and identically distributed random variables; independent of the sequence \((\zeta_t)\). Assume that the common distribution function \(F\) of \(X_1\) belongs to the Fréchet or Gumbel domains and that the tail of \((\zeta_t)\) is lighter than the tail of \((X_t)\). Using point process techniques, the authors show that the asymptotic distribution of the extreme values of the variable \(Y\) is the same as that one of the variable \(X\).

Recommendations

• scientific article; zbMATH DE number 5770651
• On extreme regression quantiles
• Test of tails based on extreme regression quantiles
• Does asymptotic linearity of the regression extend to stable domains of attraction?
• Publication:5751659

Cited in

(3)

• Mis-Specification Analysis Between Normal and Extreme Value Distributions for a Linear Regression Model
• A linear regression model with persistent level shifts: an alternative to infill asymptotics
• Repeat sampling of extreme observations: regression to the mean revisited

This page was built for publication: Asymptotic behaviour for the extreme values of a linear regression model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1769296)