This method corrects the lack of variability in conditional mean imputation (CMI) by adding an error term to the conditional mean calculation. SRI is more effective than CMI in reducing bias in the imputed values and works well with MCAR and MAR data.
Usage
stoc.impute(
data,
family = "AUTO",
tol = NULL,
robust = FALSE,
char_to_factor = FALSE,
verbose = FALSE
)Arguments
- data
a numeric matrix or data frame of at least 2 columns.
- family
the distribution family of your observations. The family arguments defaults to 'AUTO'; and it will automatically select a distribution family (gaussian, binomial, multinomial) based on the type of variable (numeric or factor). The distribution family dictates the regression model used (lm,glm, multinom). However, the user can change the family argument to match his response variable distribution and the function will adapt to this input by using the generalized linear model or beta regression.
- tol
tolerance,a numeric vector of length 1 used as multiplicative factor to standard deviation for generalized linear models. As the sample size increases, the tolerance value should be decreased to represent the decreasing variability of the sample estimate.
- robust
logical indicated whether to use robust estimation methods or ignore them. If set to 'TRUE', the function will make use of robust linear and generalized linear models to make its prediction.
- char_to_factor
transform character variable to unordered factor variable
- verbose
verbose error handling
Examples
data <- data.frame(x1 = c(stats::rnorm(87),rep(NA,13)),
x2 = stats::rnorm(100),y = stats::rnorm(100))
sri_data <- stoc.impute(data,tol = 1e-3)
summary(sri_data)
#> x1 x2 y
#> Min. :-2.809775 Min. :-2.6017 Min. :-2.21063
#> 1st Qu.:-0.557526 1st Qu.:-0.8379 1st Qu.:-0.51026
#> Median : 0.075834 Median :-0.1039 Median : 0.19840
#> Mean : 0.007388 Mean :-0.1360 Mean : 0.07453
#> 3rd Qu.: 0.583927 3rd Qu.: 0.6140 3rd Qu.: 0.72072
#> Max. : 2.430227 Max. : 2.0867 Max. : 2.69171