exactdeletion: Exact deletion residuals

In binomTools: Performing diagnostics on binomial regression models

Description

Function to derive exact values of deletion (leave-one-out) residuals for binomial regression models

Usage

exact.deletion(object)

Arguments

object

An object of class glm with a binomial family

Details

The ith deletion residual is calculated subtracting the deviances when fitting a linear logistic model to the full set of n observations and fitting the same model to a set of n-1 observations excluding the ith observation, for i = 1,...,n. This gives rise to n+1 fitting processes and may be computationally heavy for large data sets.

Approximations to the deletion residuals, as described in Williams (1987), are provided by rstudent.

Inconsistency regarding the terminology implies that the deletion residuals are called different names in the litterature, including likelihood residuals, studentized residuals, externally studentized residuals, deleted studentized residuals and jack-knife residuals. Conversely, some of these terms refer to different types of residuals

Value

A vector with exact deletion residuals

Author(s)

Merete K Hansen

References

Collett, D. (2003) Modelling binary data. Second edition. Chapman & Hall/CRC.

Fox, J. (2002) An R and S-Plus Companion to Applied Regression. Sage Publ.

Williams, D. A. (1987) Generalized linear model diagnostics using the deviance and single case deletions. Applied Statistics 36, 181-191.

See Also

Residuals, rstudent

Examples

data(beetles)
beetles.glm <- glm(cbind(y, n-y) ~ log(conc), family=binomial, data=beetles)
exact.deletion(beetles.glm)

Example output

         1          2          3          4          5          6          7 
 0.2481934  1.0856747 -0.4858414 -0.9703302 -0.4098616  0.6819076  0.3323490 
         8          9         10         11         12         13         14 
 1.1276383  1.6631429  0.5731938 -1.3209986 -1.4240437  1.3515568 -0.8309950 
        15         16 
 1.7608749  1.1673384 

binomTools documentation built on May 2, 2019, 6:53 p.m.