newtargets: Finds K nearest neighbors for new target observations

Description

Finds nearest neighbor reference observations for a given set of target observations using an established (see yai) object. Intended use is to facilitate breaking up large imputation problems (see AsciiGridImpute).

Usage

newtargets(object,newdata,k=NULL,ann=NULL)

Value

An object of class yai, that is a copy of the first argument with the following elements replaced:

call: the call.
obsDropped: a list of the row names for observations dropped for various reasons (missing data).
trgRows: a list of the row names for target observations as a subset of all observations.
xall: the X-variables for all observations.
neiDstTrgs: a matrix of distances between a target (identified by its row name) and the k references. There are k columns.
neiIdsTrgs: a matrix of reference identifications that correspond to neiDstTrgs.
neiDstRefs: set NULL as if noRefs=TRUE in the original call to yai.
neiIdsRefs: set NULL as if noRefs=TRUE in the original call to yai.
k: the value of k, replaced if changed.
ann: the value of the ann argument.

Arguments

object: an object of class yai.
newdata: a data frame or matrix of new targets for which neighbors are are found. Must include at least the X-variables used in the original call to yai.
k: if NULL, the value is taken from object, otherwise the number of nearest neighbors to find.
ann: if NULL, the value is taken from object. When TRUE ann is used to find neighbors, and when FALSE a slow exact search is used.

Author

Nicholas L. Crookston ncrookston.fs@gmail.com
Andrew O. Finley finleya@msu.edu

Examples

Run this code


require (yaImpute)

data(iris)

# set the random number seed so that example results are consistant
# normally, leave out this command
set.seed(12345)

# form some test data
refs=sample(rownames(iris),50) # just the reference observations
x <- iris[refs,1:3]  # Sepal.Length Sepal.Width Petal.Length
y <- iris[refs,4:5]  # Petal.Width Species

# build a yai object using mahalanobis
mal <- yai(x=x,y=y,method="mahalanobis")

# get imputations for the target observations (not references)
malNew <- newtargets(mal,iris[!(rownames(iris) %in% rownames(x)),])

# output a data frame of observed and imputed values for
# the observations that are not in the original yai object

impute(malNew,vars=yvars(malNew))

# in this example, Y is not specified (not required for mahalanobis).
mal2 <- yai(x=x,method="mahalanobis")
identical(foruse(mal),foruse(mal2))

if (require(randomForest))
{
  # here, method randomForest's unsupervised classification is used (no Y).
  rf <- yai(x=x,method="randomForest")
  # now get imputations for the targets in the iris data (those that are
  # not references).
  rfNew <- newtargets(rf,iris[!(rownames(iris) %in% rownames(x)),])
}