'mixError' is used to calculate the imputation error particularly in
the case of mixed-type data. Given the complete data matrix and the
data matrix containing the missing values the normalized root mean
squared error for the continuous and the proportion of falsely
classified entries for the categorical variables are computed.
Usage
mixError(ximp, xmis, xtrue)
Arguments
ximp
imputed data matrix with variables in the columns and observations in
the rows. Note there should not be any missing values.
xmis
data matrix with missing values.
xtrue
complete data matrix. Note there should not be any missing values.
Value
imputation error. In case of continuous variables only this is the
normalized root mean squared error (NRMSE, see 'help(missForest)' for
further details). In case of categorical variables onlty this is the
proportion of falsely classified entries (PFC). In case of mixed-type
variables both error measures are supplied.
Note
This function is internally used by missForest whenever a complete
data matrix is supplied.
Author(s)
Daniel J. Stekhoven, <stekhoven@stat.math.ethz.ch>
See Also
missForest
Examples
## Compute imputation error for mixed-type data:
data(iris)
## Artificially produce missing values using the 'prodNA' function:
set.seed(81)
iris.mis <- prodNA(iris, noNA = 0.2)
## Impute missing values using 'missForest':
iris.imp <- missForest(iris.mis)
## Compute the true imputation error manually:
err.imp <- mixError(iris.imp$ximp, iris.mis, iris)
err.imp