Robust regression designs when the design space consists of finitely many points

DOI10.1214/AOS/1176346406MaRDI QIDQ797941zbMATH OpenOpenAlexFDO

Authors Ker-Chau Li

Publication date 1984

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

Full work available at URL https://doi.org/10.1214/aos/1176346406

zbMATH Keywords

designs with finite supportleast squares estimatorsmaximum risk among symmetric designsnearly linear regressionrobust designsquared loss function

Mathematics Subject Classification ID

Linear regression; mixed models (62J05) Optimal statistical designs (62K05)

Review text

The author considers the model \(Y_ i=\alpha +\beta x_ i+g(x_ i)+\epsilon_ i\), \(i=1,2,...,n\), where \(\{\epsilon_ i\}\) are uncorrelated random variables with mean 0 and variance \(\sigma^ 2\), \(x_ i\in X=\{k/2N,-k/2N|\quad k=1,2,...,N\}\) for some fixed positive integer \(N\geq 3\) and \[ g\in G=\{h|\quad | h(x)|\leq \delta,\quad\min_{\alpha,\beta}\int (h(x)-\alpha -\beta x)^ 2d\mu (x)=\int h^ 2(x)d\mu (x)\}, \] with \(\delta \geq 0\) and \(\mu\) the uniform probability measure on X. The loss function \(\omega_ 1(\alpha - {\hat\alpha })^ 2+\omega_ 2(\beta -{\hat\beta })^ 2\) is assumed where \(\omega_ 1,\omega_ 2\) are specified nonnegative constants and \({\hat\alpha }\),\({\hat\beta }\) are the least squares estimators of \(\alpha\) and \(\beta\), respectively. Designs symmetric about zero are obtained which minimize the expected loss.

Recommendations

• Some model robust designs in regression
• Robust designs and optimality of least squares for regression problems
• Robust designs for approximately linear regression: \(M\)-estimated parameters
• scientific article; zbMATH DE number 1532007
• Minimax designs for approximately linear regression

Cited in

(11)

• Model robust extrapolation designs
• Optimal designs for regression models with possible bias
• Minimax designs for approximately linear regression
• Robust designs for approximately linear regression: \(M\)-estimated parameters
• Robust designs for approximately polynomial regression
• Minimax regression designs for approximately linear models with autocorrelated errors
• Restricted optimal designs for linear and quadratic polynomial regressions
• INTEGER-VALUED, MINIMAX ROBUST DESIGNS FOR APPROXIMATELY LINEAR MODELS WITH CORRELATED ERRORS
• scientific article; zbMATH DE number 862313 (Why is no real title available?)
• Minimax designs for linear regression models with bias in a reproducing kernel Hilbert space in a discrete set
• Minimax designs for misspecified linear models with L_-norm bounded departures

This page was built for publication: Robust regression designs when the design space consists of finitely many points

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q797941)