WEPOR: A Weighted Polynomial Regression Program

WEPOR: A Weighted Polynomial Regression Program

Two weighted least squares regression programs are documented and exemplified. Both programs apply to polynomial models with one independent variable. Program WEPOR was formulated for the case in which the error terms have different variances but are uncorrelated, whereas WEPOR2 deals with the probl...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Shields, Patricia A, Thomas, Marlin A
Format: Report
Sprache:eng
Schlagworte:
ANALYSIS OF VARIANCE
ERRORS
ESTIMATES
LEAST SQUARES METHOD
LINEAR REGRESSION ANALYSIS
MATHEMATICAL MODELS
METHODOLOGY
POLYNOMIALS
Statistics and Probability
VARIABLES
WEIGHTING FUNCTIONS
WEPOR(WEIGHTED POLYNOMIAL REGRESSION)
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Two weighted least squares regression programs are documented and exemplified. Both programs apply to polynomial models with one independent variable. Program WEPOR was formulated for the case in which the error terms have different variances but are uncorrelated, whereas WEPOR2 deals with the problem of different variances and correlated error terms. Output from both programs include ANOVA (ANalysis Of VAriance) tables, predicted values of the dependent variable and the associated residuals, confidence and prediction limits for selected synthetic points, and a plot of the sample points, the regression curve, and the confidence and prediction limits.