buildConsensus: Finds the consensus imputations among a list of yai objects

In yaImpute: Nearest Neighbor Observation Imputation and Evaluation Tools

Finds the consensus imputations among a list of yai objects

Description

Several objects of class yai are combined into a new object forming a consensus among the many. The intention is that the many would be formed by running yai several times with bootstrap=TRUE or by varying other options.

Usage

buildConsensus(reps, noTrgs=FALSE, noRefs=FALSE, k=NULL)

Arguments

reps	a list of objects class yai.
noTrgs	If TRUE neighbor relationships for target observations are not merged.
noRefs	If TRUE neighbor relationships for reference observations are not merged.
k	If not specified, the minimum value of k among the objects is used.

Value

An object of class yai

Author(s)

Nicholas L. Crookston [e-mail]
John Coulston [e-mail]

See Also

yai

Examples

require (yaImpute)
data(iris)

set.seed(123) 

# form some test data, y's are defined only for reference
# observations.
refs=sample(rownames(iris),50)
x <- iris[,1:2]      # Sepal.Length Sepal.Width
y <- iris[refs,3:4]  # Petal.Length Petal.Width

reps <- replicate(20, yai(x=x,y=y,method="msn",bootstrap=TRUE,k=2),
                  simplify=FALSE)

buildConsensus(reps)

yaImpute documentation built on Nov. 4, 2022, 1:06 a.m.