Points of Significance: Regression diagnostics

So far in our discussion of linear regression, we have seen that the estimated regression coefficients and predicted values can be difficult to interpret1. When the predictors are correlated2, the magnitude and even the sign of the estimated regression coefficients can be highly variable, although t...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Nature methods Jg. 13; H. 5; S. 385
Hauptverfasser: Altman, Naomi, Krzywinski, Martin
Format: Journal Article
Sprache:Englisch
Veröffentlicht: New York Nature Publishing Group 01.05.2016
Schlagworte:
ISSN:1548-7091, 1548-7105
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:So far in our discussion of linear regression, we have seen that the estimated regression coefficients and predicted values can be difficult to interpret1. When the predictors are correlated2, the magnitude and even the sign of the estimated regression coefficients can be highly variable, although the predicted values may be stable. When outliers are present3, both the estimated regression coefficients and the predicted values can be influenced. This month, we discuss diagnostics for the robustness of the estimates and of the statistical inferencethat is, the t-tests, confidence intervals and prediction intervals that are computed on the basis of assumptions that the errors are additive, normal and independent and have zero mean and constant variance.
Bibliographie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:1548-7091
1548-7105
DOI:10.1038/nmeth.3854