impRZilr: EM-based replacement of rounded zeros in compositional data

Description

Parametric replacement of rounded zeros for compositional data using classical and robust methods based on ilr-transformations with special choice of balances.

Usage

impRZilr(x, maxit = 10, eps = 0.1, method = "roundedZero", dl = rep(0.05, ncol(x)), bruteforce = FALSE)

Arguments

data.frame or matrix

maxit

maximum number of iterations

eps

convergency criteria

method

either roundedZeor or roundedZeroRobust

Detection limit for each variable

bruteforce

sets imputed values above the detection limit to the detection limit. Replacement above the detection limit are only exeptionally occur due to numerical instabilities. The default is FALSE!

Value

xOrigOriginal data frame or matrix
xImpImputed data
windIndex of the missing values in the data
iterNumber of iterations
epseps

Details

Statistical analysis of compositional data including zeros runs into problems, because log-ratios cannot be applied. Usually, rounded zeros are considerer as missing not at random missing values.

The algorithm iteratively imputes parts with rounded zeros whereas in each step (1) an specific ilr transformation is applied (2) tobit regression is applied (3) the rounded zeros are replaced by the expected values (4) the corresponding inverse ilr transformation is applied. After all parts are imputed, the algorithm starts again until the imputations do not change.

Examples

Run this code

data(arcticLake)
x <- arcticLake
## generate rounded zeros artificially:
x[x[,1] < 5, 1] <- 0
x[x[,2] < 47, 2] <- 0
xia <- impRZilr(x, dl=c(5,47,9999), eps=0.05)
xia$xImp