marginalCompleter: Wrapper to complete a dataset based on (univariate) marginal...

In EMLasso: Fit (logistic) LASSO when there is missing data in the predictors

Description

Wrapper to complete a dataset based on (univariate) marginal distributions

Usage

  marginalCompleter(weightsName = "weights",
    orgriName = "orgri", maxFullNACatCols = 6,
    howManyIfTooMany = 1000)

  completeMarginal(object,ds, out, rowsToUse, verbosity=0)

  ## S3 method for class 'function'
completeMarginal(object,ds, out,
    rowsToUse, verbosity=0)

  ## Default S3 method:
completeMarginal(object,ds, out,
    rowsToUse, verbosity=0)

Arguments

weightsName	object that holds information to convert an imputed dataset like dfrConversionProps or a custom implementation.
orgriName	dataset for which the lambdas need to be found
maxFullNACatCols	see rCatsAndCntInDfr
howManyIfTooMany	see rCatsAndCntInDfr
object	object/function that will perform the actual imputation
ds	dataset that needs completion
out	outcomes
rowsToUse	rowindices of the rows that can actually be used
verbosity	The higher this value, the more levels of progress and debug information is displayed (note: in R for Windows, turn off buffered output)

Value

function that can actually perform the conversion – used in EMLasso.1l

object of class marginalCompleted, holding items:

1. ds: completed dataset

2. out: outcome variable

3. weights: weights to apply to each row of ds

4. completedUse: rowindices within ds to be actually used

Note

abstracts away creation of marginal completion: see EMLasso.1l

See Also

EMLasso.1l

EMLasso.1l

EMLasso.1l

EMLasso documentation built on May 2, 2019, 5:49 p.m.