Fraction of variance unexplained
From Wikipedia, the free encyclopedia
This article relies largely or entirely on a single source. Relevant discussion may be found on the talk page. Please help improve this article by introducing citations to additional sources.
Find sources: "Fraction of variance unexplained" – news ·newspapers · books · scholar · JSTOR (June 2020)
For broader coverage of this topic, see Explained variation.
In statistics, the fraction of variance unexplained (FVU) in the context of a regression task is the fraction of variance of the regressand (dependent variable) Y which cannot be explained, i.e., which is not correctly predicted, by the explanatory variables X.
1Formal definition
3See also
Formal definition​[​edit​]
Suppose we are given a regression function yielding for each
an estimate
is the vector of the ith observations on all the explanatory variables.​[1]​:​181 We define the fraction of variance unexplained (FVU) as:
where R2 is the coefficient of determination and VARerr and VARtot are the variance of the residuals and the sample variance of the dependent variable. SSerr (the sum of squared predictions errors, equivalently the residual sum of squares​), SStot (the total sum of squares​), and SSreg (the sum of squares of the regression, equivalently the explained sum of squares) are given by
Alternatively, the fraction of variance unexplained can be defined as follows:
where MSE(f) is the mean squared error of the regression function ƒ.
It is useful to consider the second definition to understand FVU. When trying to predict Y, the most naïve regression function that we can think of is the constant function predicting the mean of Y, i.e.,
. It follows that the MSE of this function equals the variance of Y; that is, SSerr = SStot, and SSreg = 0. In this case, no variation in Y can be accounted for, and the FVU then has its maximum value of 1.
More generally, the FVU will be 1 if the explanatory variables X tell us nothing about Y in the sense that the predicted values of Y do not covary with Y. But as prediction gets better and the MSE can be reduced, the FVU goes down. In the case of perfect prediction where
for all i, the MSE is 0, SSerr = 0, SSreg = SStot, and the FVU is 0.
See also​[​edit​]
^ Achen, C. H. (1990). "'What Does "Explained Variance" Explain?: Reply". Political Analysis. 2 (1): 173–184. doi​:​10.1093/pan/2.1.173​.
Categories: Parametric statisticsStatistical ratiosLeast squares
This page was last edited on 15 April 2021, at 18:12 (UTC).
Text is available under the Creative Commons Attribution-ShareAlike License; additional terms may apply. By using this site, you agree to the Terms of Use and Privacy Policy. Wikipedia® is a registered trademark of the Wikimedia Foundation, Inc., a non-profit organization.
Privacy policy
About Wikipedia
Contact Wikipedia
Mobile view
Cookie statement
TalkContributionsCreate accountLog in
ReadEditView history
Visit the main pageMain pageContentsCurrent eventsRandom articleAbout WikipediaContact usDonateHelpLearn to editCommunity portalRecent changesUpload fileWhat links hereRelated changesUpload fileSpecial pagesPermanent linkPage informationCite this pageWikidata itemDownload as PDFPrintable versionAdd links