impStand: Standardize Imputations

In NINAnor/NIcalc: Functions to Calculate a Nature Index

Standardize Imputations

Description

Optional *post hoc* adjustment of sets of multiple imputations for a set of missing indicator observations.

Usage

impStand(x = NULL, imputations = NULL)

Arguments

x	list of class niInput.
imputations	list of class niImputations

Details

In some instances, especially when working with very unbalanced data, the imputation model used in imputeData may not produce neutral imputations as desired. In such cases, *post hoc* adjustment of imputed values may be used to avoid that imputations introduce new patterns in the data set. In this context, neutral values are imputations with an expected deviation from the (grand) mean equal to the sum of 1) the mean deviation for observations from the year in question, and 2) the mean deviation from annual means for observations of the indicator in question.

Value

A list of class niImputations

Author(s)

Bård Pedersen

See Also

imputeData, imputeDiagnostics, mice::mice.
The vignette objectsInNIcalc gives a detailed description of niImputations and niInput lists.

Examples

## Not run: 
imputedValues <- imputeData(x = themeData,
                            nSim = 1000,
                            transConst = 0.01,
                            maxit = 20,
                            printFlag = TRUE)
imputedValues <- impStand(x = themeData,
                          imputations = imputedValues)

## End(Not run)

NINAnor/NIcalc documentation built on Oct. 26, 2023, 9:37 a.m.