mi.info {mi}R Documentation

Function to create information matrix for missing data imputation

Description

Produces matrix of information needed to impute the missing data. After the information is extracted user has the option of changing the default.

Usage

  mi.info(data, threshhold  = 0.99999)
  ## S4 method for signature 'mi.info':
  print(x, ...)
  ## S4 method for signature 'mi.info':
  show(object)

Arguments

data dataframe or matrix of dataset with missing data coded as NAs.
threshhold Threshhold value for correlation to be considered a problem.
x An object of a class '"mi.info"'.
object An object of a class '"mi.info"'.
... Currently not used.

Details

~~ If necessary, more details than the description above ~~

Value

info information matrix
-name: Name of variable
-imp.order: Imputation Order
-nmis: Number of missing
-type: Type of variable
-var.class: Class of input variable
-level: Levels in the input varialbe
-include: Include in the imputation process or not
-is.ID: Is ID variable or not
-all.missing: All observation missing or not
-correlated: Correlated variables
-determ.pred: Deterministic predictor
-imp.formula: Imputation formula
-params: Parameters for the imputation model
-other: Currently not used

Author(s)

Masanao Yajima yajima@stat.columbia.edu, M.Grazia Pittau grazia@stat.columbia.edu, Andrew Gelman gelman@stat.columbia.edu

References

Andrew Gelman and M. Grazia Pittau, A flexible program for missing-data imputation and model checking, Technical report, Columbia University, New York; Andrew Gelman and Jennifer Hill, Data Analysis Using Regression and Multilevel/Hierarchical Models, Cambridge University Press, 2007.

See Also

mi

Examples

  data(CHAIN)
  info.CHAIN <- mi.info(CHAIN)
  
  info.CHAIN$imp.order # imputation order
  
  info.CHAIN$imp.formula # imputation formula
  info.CHAIN[["age.W1"]]$imp.formula  #imputation formula for specific variable

[Package mi version 0.04-6 Index]