Multiple imputations object

DESCRIPTION:

This is an list of class miList whose elements correspond to multiple imputations.

GENERATION:

The generator function for this class of objects is miList. miList objects are commonly created as the output of miEval and miApply, functions that perform calculations on multiple imputations.

METHODS:

The "miList" class of objects has associated methods: anyMissing.miList, as.miVariable.miList, data.frameAux.miList, miDiscard.miList, miTrim.miList, numberMissing.miList, and print.miList.

Other functions should be used within , e.g. miEval(summary(x)).

STRUCTURE:

An miList is a list with class "miList". Each component of this list corresponds to one set of (multiple) imputation; there must be at least two imputations.

DETAILS:

This is a very general class. Each component of the list corresponds to a single set of multiple imputations; the components typically have the same structure but different numerical values.

For example, if a data frame contains M sets of multiple imputations, then the output of miEval for a linear regression using that data will be an miList with M components, each of which is an lm object obtained by replacing missing values in the data frame with one of the M sets of imputations, then running the regression.

SEE ALSO:

, , , , .

EXAMPLES:

x <- miList(list(3, 4, 5)) 
x           # Default names are 1, 2, ... 
oldClass(x) # has old-style class "miList"