Measurement Error in Regression Analysis.

Abstract

Consider the linear regression model Y = x theta + epsilon where Y denotes a vector of n observations on the dependent variable, x is a known matrix, theta is a vector of parameters to be estimated and epsilon is a random vector of uncorrelated errors. If X'X is nearly singular, that is if the smallest characteristic root of X'X is small then a small perturbation in the elements of X, such as due to measurement errors, induces considerable variation in the least squares estimate of theta. In this paper we examine for the asymptotic case when n is large the effect of perturbation with regard to the bias and mean squared error fo the estimate. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 08, 1978
Accession Number
ADA087287

Entities

People

  • Amitava Mitra
  • Khursheed Alam

Organizations

  • Clemson University

Tags

DTIC Thesaurus Topics

  • Computer Programs
  • Contracts
  • Data Sets
  • Estimators
  • Mathematical Analysis
  • Mathematics
  • Measurement
  • Military Research
  • Monte Carlo Method
  • Numerical Analysis
  • Perturbations
  • Random Variables
  • Regression Analysis
  • Simulations
  • Statistical Algorithms
  • Universities

Fields of Study

  • Mathematics

Readers

  • Analytical Mechanics
  • Atmospheric Science / Meteorology, specifically Wind Wave Turbulence.
  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)