Robust spectral regression

DOI10.1214/AOS/1176350255MaRDI QIDQ1102682zbMATH OpenOpenAlexFDO

Authors Alexander M. Samarov

Publication date 1987

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1214/aos/1176350255

zbMATH Keywords

efficiency robustnessfrequency domaingeneralized least squaresregression estimationregression spectrumsecond-order stationary processserial correlation correctionspectral densityweighted least squares

Mathematics Subject Classification ID

Robustness and adaptive procedures (parametric inference) (62F35) Linear regression; mixed models (62J05) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Inference from stochastic processes and spectral analysis (62M15)

Review text

Let \(y_ t=x'_ t\beta +u_ t\), \(t\in Z^ 1\), where the disturbances \(u_ t\) follow a second-order stationary process with E \(u_ t=0\). The spectral density f(\(\lambda)\) of a process \(u_ t\) is unknown but belongs to a neighborhood of some special spectral density \(f_ 0(\lambda)\). Using methods developed in robust statistics, the problem of the efficiency of linear (in the y's) estimators of \(\beta\) is considered.

Recommendations

• On the robust estimation of the regression coefficients of random processes with singular spectrum
• Robustness of the autoregressive spectral estimate for linear processes with infinite variance
• Frequency domain analysis of robust signal estimators
• scientific article; zbMATH DE number 3885166
• scientific article; zbMATH DE number 3905646

Cited in

(6)

• Frequency domain analysis of robust signal estimators
• Minimax regression designs for approximately linear models with autocorrelated errors
• A theory of robust long-run variance estimation
• Regresion espectral sesgada
• One‐step M‐estimators in the linear model, with dependent errors
• Analyzing the impact of regularization on REMSE

This page was built for publication: Robust spectral regression

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1102682)