scientific article; zbMATH DE number 3864319
Consider the linear model \(y_ i=\beta_ 1X_{i1}+...+\beta_ px_{ip}+e_ i (i=1,2,...)\) where \(x_{ij}\) are known constants, \(\beta_ 1,...,\beta_ p\) are unknown parameters, and \(e_ i\) are random errors. Denote \(\beta =(\beta_ 1,...,\beta_ p)^ t\), \(X_ n=(x_{ij})_{n\times p}\). Let \(b_ n=(b_{n1},...,b_{np})^ t\) be the least squares estimate of \(\beta\) based on the design matrix \(X_ n\) and \((y_ 1,...,y_ n)^ t.\) In this paper, a sufficient condition on \(\lim_{n\to \infty} E\| b_ n-\beta \|^ r=0\) for some \(r\in(0,2]\) is given, and convergence rates are obtained. These results improve those of \textit{X. Chen}, Acta Math. Sin. 24, 36-44 (1981; Zbl 0473.62051).
- Some results on consistency of LS estimates
- Studies on consistency of LSE in China
- Consistency of least squares estimates in a system of linear correlation models
- scientific article; zbMATH DE number 3878185 (Why is no real title available?)
- scientific article; zbMATH DE number 3911512 (Why is no real title available?)
- scientific article; zbMATH DE number 3986459 (Why is no real title available?)
- scientific article; zbMATH DE number 22267 (Why is no real title available?)
- scientific article; zbMATH DE number 510466 (Why is no real title available?)
- scientific article; zbMATH DE number 617148 (Why is no real title available?)
- scientific article; zbMATH DE number 2112978 (Why is no real title available?)
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3332098)