simul_incomplete: Toy multilevel categorical dataset with missing entries...

In davidevdt/BMLCimpute: BMLCimpute: Bayesian Multilevel Latent Class Models for the Multiple Imputation of Nested Categorical Data

Description

The dataset contains 1000 level-1 artificial observations grouped into 200 level-2 groups. The dataset consists of: one identifier for the group, one identifier for the level-1 units, six level-1 binary variables (5 predictors and 1 response), five level-2 binary variables. The level-1 variables X1, X2, X5 and the level-2 variables Z1 and Z3 have missing observations, generated through a MAR mechanism. The dataset and the missing data were generated following the simulation study settings in Vidotto, Vermunt, van Deun (2018). Dataset used for testing of multilevel imputation models; model performance can be assessed by using the simul dataset.

Usage

data(simul_incomplete)

Format

A data frame with 13 columns and 1,000 rows (level-1 units) from 200 groups (level-2 units):

[,1]	GroupID	numeric	Level-2 unit Identifier
[,2]	UnitID	numeric	Level-1 unit Identifier
[,3]	X1,...,X5	binary	Level-1 predictors
[,4]	Z1,...,Z5	binary	Level-2 predictors
[,5]	Y	binary	Response Variable

References

• [1] Vidotto D., Vermunt J.K., Van Deun K. (2018). 'Bayesian Multilevel Latent Class Models for the Multiple Imputation of Nested Categorical Data'. Journal of Educational and Beahvioral Statistics 43(5), 511-539.

Examples

## Not run: 

library(BMLCimpute)

data(simul_incomplete)


## End(Not run)

davidevdt/BMLCimpute documentation built on June 5, 2019, 12:36 a.m.