A Comparison Between Maximum Likelihood and Generalized Least Squares in a Heteroscedastic Linear Model

DOI10.2307/2287321zbMath0514.62079OpenAlexW4247521989MaRDI QIDQ3660716

David Ruppert, Raymond J. Carroll

Publication date: 1982

Full work available at URL: http://www.lib.ncsu.edu/resolver/1840.4/3507

zbMATH Keywords

comparisonrobustnessmaximum likelihoodcontiguitygeneralized least squaresweighted least squaresheteroscedastic linear modelfunctional relationshiphypothesis of homoscedasticitysmall misspecifications

Mathematics Subject Classification ID

Robustness and adaptive procedures (parametric inference) (62F35) Analysis of variance and covariance (ANOVA) (62J10)

Related Items

Iterated weighted least squares in heteroscedastic lineaipmod%81è, Robust methods for heteroskedastic regression, Two-stage generalized moment method with applications to regressions with heteroscedasticity of unknown form, Testing goodness-of-fit for nonlinear regression models with heterogeneous variances, Estimation in one–way classification with heteroscedastic error, Nonlinear factor analysis as a statistical method., Metamodeling and robust minimization approach for the identification of elastic properties of composites by vibration method, A Model for a Proportional Treatment Effect on Disease Progression, Covariance matrices of S robust regression estimators, Robust tests in nonlinear regression models, Two-Step Estimation in a Heteroscedastic Linear Regression Model, Robust estimation in certain heteroscedastic linear models when there are many parameters, Quasi-likelihood estimation for GLM with random scales, Model misspecification in parametric dual modeling