Function_name | Package_name | Use |
---|---|---|
aggr | VIM | To calculate/plot the missing values in the variables |
boxplot | base/graphics | To produce a box plot |
bwplot | mice | To produce box plot to compare the imputed and observed values |
colMeans | base | To compute the column-wise mean, i.e., mean for each variable/column |
complete | mice | To extract the imputed dataset |
complete.cases | base/stats | To select the complete cases, i.e., observations without missing values |
D1 | mice | To conduct the multivariate Wald test with D1-statistic |
densityplot | mice | To produce desnsity plots |
expression | base | To set/create an expression |
imputationList | mice | To combine multiple imputed datasets |
marginplot | VIM | To draw a scatterplot with additional information when there are missing values |
mcar_test | naniar | To conduct Little's MCAR test |
md.pattern | mice | To see the pattern of the missing data |
mice | mice | To impute missing data where the argument m represents the number of multiple imputation |
MIcombine | mitools | To combine/pool the results using Rubin's rule |
MIextract | mitools | To extract parameters from a list of outputs |
na.test | misty | To conduct Little's MCAR test |
parlmice | mice | To run `mice` function in parallel, i.e., parallel computing of mice |
plot_missing | DataExplorer | To plot the profile of missing values, e.g., the percentage of missing per variable |
pool | mice | To pool the results using Rubin's rule |
pool.compare | mice | To compare two nested models |
pool_mi | miceadds | To combine/pool the results using Rubin's rule |
quickpred | mice | To set imputation model based on the correlation |
sim_slopes | interactions | To perform simple slope analyses |
TestMCARNormality | MissMech | To test multivariate normality and homoscedasticity in the context of missing data |
unlist | base | To convert a list to a vector |
R functions (M)
The list of new R functions introduced in this Missing data analysis lab component are below: