mi.preprocess {mi}R Documentation

Preproessing and Postprocessing mi data object

Description

Function for propressing and postprocessing nonnegative, and positive-continuous variable types in mi data object

Usage

  mi.preprocess(data, info)
  mi.postprocess(mi.data, info)

Arguments

data the data.frame to be imputed.
info the information matrix, see mi.info.
mi.data the imputed data list, obtained from mi.completed

Details

mi.proprocess will transform the nonnegative and positive-continuous variable types. If the variable is of nonnegative type, the function transforms the variable into two variables: an indicator indicates whether the value is postive or not and a transformed variable that takes on all positive value and is transformed either by taking a log; 0 and NA will be treated as missing for such a variable. If the variable is of positive-continuous type, it will be transformed by taking a log.

mi.postprocess will transform the imputed dataset back to its original form. The imputed dataset is obtained from mi.completed function.

Value

a data.frame or a list of dataframe.

Author(s)

Yu-Sung Su ys463@columbia.edu, Andrew Gelman gelman@stat.columbia.edu

References

Yu-Sung Su, Andrew Gelman, Jennifer Hill, Masanao Yajima. Forthcoming. “Multiple Imputation with Diagnostics (mi) in R: Opening Windows into the Black Box”. Journal of Statistical Software.

See Also

mi.completed


[Package mi version 0.08-06 Index]