bigpca (version 1.1)

subcor.select: Selection of the most correlated variable subset

Description

Returns a subset (size='keep') of row or column numbers that are most correlated to other variables in the dataset (or if hi.cor=F), then those that are least correlated. This function performs cor() on a small subset of columns and all rows (when rows=TRUE, or vice -versa when rows=FALSE), and selects rows (rows=TRUE) with greatest/least absolute sum of correlations.

Usage

subcor.select(bigMat, keep = 0.05, rows = TRUE, hi.cor = TRUE,
  dir = getwd(), random = TRUE, ram.gb = 0.1)

Arguments

bigMat

a big.matrix, matrix or any object accepted by get.big.matrix()

keep

numeric, by default a proportion (decimal) of the original number of rows/columns to choose for the subset. Otherwise if an integer>2 then will assume this is the size of the desired subset, e.g, for a dataset with 10,000 rows where you want a subset size of 1,000 you could set 'keep' as either 0.1 or 1000.

rows

logical, whether the subset should be of the rows of bigMat. If rows=FALSE, then the subset is chosen from columns, would be equivalent to calling subpc.select(t(bigMat)), but avoids actually performing the transpose which can save time for large matrices.

hi.cor

logical, whether to choose the most correlated (TRUE) or least correlated subset (FALSE).

dir

the directory containing the bigMat backing file (e.g, parameter for get.big.matrix()).

random

logical, passed to uniform.select(), whether to take a random or uniform selection of columns (or rows if rows=FALSE) to run the subset PCA.

ram.gb

maximum size of the matrix in gigabytes for the subset PCA, 0.1GB is the default which should result in minimal processing time on a typical system. Increasing this increases the processing time, but also the representativeness of the subset chosen. Note that some very large matrices will not be able to be processed by this function unless this parameter is increased; basically if the dimension being thinned is more than 5 this memory limit (see estimate.memory() from NCmisc).

Value

A set of row or column indexes (depents on 'rows' parameter) of the most inter-correlated (or least) variables in the matrix.

See Also

thin, uniform.select, get.big.matrix

Examples

# NOT RUN {
mat <- matrix(rnorm(200*2000),ncol=200)
bmat <- as.big.matrix(mat)
ii1 <- subcor.select(bmat,.05,rows=TRUE) # thin down to 5% of the rows
ii2 <- subcor.select(bmat,45,rows=FALSE) # thin down to 45 columns
prv(ii1,ii2)
# show that rows=T is equivalent to rows=F of the transpose (random must be FALSE)
ii1 <- subcor.select(mat,.4,rows=TRUE,random=FALSE)
ii2 <- subcor.select(t(mat),.4,rows=FALSE,random=FALSE)
print(all.equal(ii1,ii2))
# }