A Semi‐parametric Regression Model with Errors in Variables (Q4828210)

From MaRDI portal

!

This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:

scientific article; zbMATH DE number 2118816
Language Label Description Also known as
default for all languages
No label defined
    English
    A Semi‐parametric Regression Model with Errors in Variables
    scientific article; zbMATH DE number 2118816

      Statements

      A Semi‐parametric Regression Model with Errors in Variables (English)
      0 references
      0 references
      0 references
      24 November 2004
      0 references
      semi-parametric regression
      0 references
      errors in variables
      0 references
      deconvolution kernel estimator
      0 references
      asymptotic normality
      0 references
      convergence rate
      0 references
      The model under consideration is a semi-parametric regression model such that the conditional expectation of the univariate \(Y\) is \(E(Y\mid t_0,X_0)=X_0^T (\theta)+g(t_0)\), where \(X_0\) and \(t_0\) are \(p\)- and 1-dimensional, respectively, \(\theta\) is an \(p\times1\) unknown parameter vector, \(g(\cdot)\) is an unknown function, and both \(X_0\) and \(t_0\) are measured with errors, i.e., \(X=X_0+u\), \(t=t_0+v\). The authors propose a combination of the method of moments and deconvolution to construct a new class of parametric estimates together with a nonparametric kernel estimator. Strong convergence, optimal rate of weak convergence and asymptotic normality are proved, and simulation study results are presented.
      0 references

      Identifiers

      0 references
      0 references
      0 references
      0 references
      0 references
      0 references