imir: Imputation Indicator for Responded Items IMIR

In sdap: Statistical Data Preparation Indicators

Description

Version of imi() for responded items. The aggregated imputation impact for variable groups allows measuring the effect of imputations of values with original response on the distribution of categorical variables.

Usage

imir(data, r1ij, bij, gij, obsi = 1:nrow(gij), varj = 1:ncol(gij), weight)

Arguments

data	(mandatory): A dataframe containing the data to be processed.
r1ij	(mandatory): A matrix containing the response indicators for a given dataframe.
bij	(optional): A matrix containing the structurally missingness indicators. bij can be calculated using the function smind. If the argument bij is missing, the indicator is calculated without considering a missingness indicators matrix.
gij	(mandatory): A matrix containing the imputation indicators for a given dataframe. gij can be calculated using the function impind.
obsi	(optional): A vector with the observations in rij to be processed. If the argument obs is missing, all observations are processed.
varj	(optional): A vector with the variables (column numbers) to be considered for the calculation. If the argument varj is missing, all variables are considered for the indicator.
weight	(optional): A vector of weights to be considered when calculating the indicator. If no weight vector is given as an argument, the indicator is calculated without considering different weights.

Value

A list with the following elements: variables (variables), observations (observations), Number of imputations detected (imputations), Indicator IMIR (imir).

Author(s)

Beat Hulliger - Juan Berdugo

sdap documentation built on May 2, 2019, 6:52 p.m.