Journal of the American Statistical Association · 1990 · 1.1K citations · 0 references
BusinessBiostatisticsEconometricsStatistical ModelingStatisticsData TransformationGeneralized Linear ModelsRegressionRobust Statistical ModelingRobust StatisticsVariance AnalysisEstimation Of Regression ModelsGeneralized Least Squares EstimationHeteroscedastic Error ModelsHeteroscedasticity ModelingHeteroskedasticity ModelingRobust RegressionWeighted Least SquaresWeighted Least Squares Estimation
The paper reviews generalized least squares, heteroscedasticity analysis, variance function estimation, and the integration of transformations and weighting, highlighting influence, robustness, technical aspects, and remaining open problems. The study introduces the Transform‑Both‑Sides methodology for regression analysis. The article contains an introduction, references, and an index.
Introduction. Generalized Least Squares and the Analysis of Heteroscedasticity. Estimation and Inference for Variance Functions. The Transform-Both-Sides Methodology. Combining Transformations and Weighting. Influence and Robustness. Technical Complements. Some Open Problems. References. Index.