= Cook's Distance =

'''Cook's distance''' is a measure of an observation's influence in an estimated model.

<<TableOfContents>>

----



== Description ==

Cook's distance for observation ''i'' is calculated as:

{{attachment:cook.svg}}

where ''k + 1'' is the number of parameters fit (including the intercept), ''r,,i,,'' is the case's [[Statistics/Residuals#Standardized_Residuals|standardized residual]], and ''h,,i,i,,'' is the case's [[Statistics/Leverage#Hat_Matrix|hat value]].

If a high level of influence for an observation is inappropriate, it can be considered an '''outlier''' and may be excluded from estimation. A common rule of thumb is to consider any case with a Cook's distance greater than 1 an outlier. 



----
CategoryRicottone