MICE: multivariate imputation by chained equations (normal model)

van Buuren's Algorithm 4.3 with the Bayesian normal linear imputation of Algorithm 3.1 (mice method "norm"): random-draw starting imputations, then per iteration each incomplete variable is imputed from a posterior draw of its regression on the other (currently complete) variables.

Usage

morie_miord2(data, m = 5, maxit = 5, seed = 0, kappa = 1e-04)

Arguments

data: Numeric matrix or data frame; missing entries NA.
m: Number of imputed data sets.
maxit: Iterations per chain.
seed: SplitMix64 seed (mirrors the Python arm).
kappa: Ridge parameter of Algorithm 3.1.

Value

A list with elements imputations (list of m completed matrices), missing_mask, m, maxit, column_means, seed, method.

References

van Buuren, S. (2018). Flexible Imputation of Missing Data, 2nd ed. Chapman & Hall/CRC. van Buuren, S. and Groothuis-Oudshoorn, K. (2011). JSS, 45(3). Rubin, D. B. (1987). Multiple Imputation for Nonresponse in Surveys. Wiley.

Examples

V <- c(1, 2, 3, 4, 5, 6, 7, 8)
morie_miord2(V)
#> $imputations
#> $imputations[[1]]
#>      [,1]
#> [1,]    1
#> [2,]    2
#> [3,]    3
#> [4,]    4
#> [5,]    5
#> [6,]    6
#> [7,]    7
#> [8,]    8
#> 
#> $imputations[[2]]
#>      [,1]
#> [1,]    1
#> [2,]    2
#> [3,]    3
#> [4,]    4
#> [5,]    5
#> [6,]    6
#> [7,]    7
#> [8,]    8
#> 
#> $imputations[[3]]
#>      [,1]
#> [1,]    1
#> [2,]    2
#> [3,]    3
#> [4,]    4
#> [5,]    5
#> [6,]    6
#> [7,]    7
#> [8,]    8
#> 
#> $imputations[[4]]
#>      [,1]
#> [1,]    1
#> [2,]    2
#> [3,]    3
#> [4,]    4
#> [5,]    5
#> [6,]    6
#> [7,]    7
#> [8,]    8
#> 
#> $imputations[[5]]
#>      [,1]
#> [1,]    1
#> [2,]    2
#> [3,]    3
#> [4,]    4
#> [5,]    5
#> [6,]    6
#> [7,]    7
#> [8,]    8
#> 
#> 
#> $missing_mask
#>       [,1]
#> [1,] FALSE
#> [2,] FALSE
#> [3,] FALSE
#> [4,] FALSE
#> [5,] FALSE
#> [6,] FALSE
#> [7,] FALSE
#> [8,] FALSE
#> 
#> $m
#> [1] 5
#> 
#> $maxit
#> [1] 5
#> 
#> $column_means
#> $column_means[[1]]
#> [1] 4.5
#> 
#> $column_means[[2]]
#> [1] 4.5
#> 
#> $column_means[[3]]
#> [1] 4.5
#> 
#> $column_means[[4]]
#> [1] 4.5
#> 
#> $column_means[[5]]
#> [1] 4.5
#> 
#> 
#> $seed
#> [1] 0
#> 
#> $method
#> [1] "MICE norm (van Buuren Algs. 3.1 + 4.3)"
#>