Asymptotic properties of covariate-adjusted regression with correlated errors

Damla Şentürk, Danh V. Nguyen

Research output: Contribution to journalArticle

6 Scopus citations

Abstract

In covariate-adjusted regression (CAR), the response (Y) and predictors (Xr, r = 1, ..., p) are not observed directly. Estimation is based on n independent observations {over(Yi, ̃), over(X, ̃)r i, Ui}i = 1 n, where over(Y, ̃)i = ψ (Ui) Yi, over(X, ̃)r i = φ{symbol}r (Ui) Xr i and ψ ({dot operator}) and {φ{symbol}r ({dot operator})}r = 1 p are unknown functions. In this paper, we discuss the asymptotic properties of this method when the observations are correlated, as in regression models for repeated measurements.

Original languageEnglish (US)
Pages (from-to)1175-1180
Number of pages6
JournalStatistics and Probability Letters
Volume79
Issue number9
DOIs
StatePublished - May 1 2009

    Fingerprint

ASJC Scopus subject areas

  • Statistics, Probability and Uncertainty
  • Statistics and Probability

Cite this