Though similar to est.nearneighbor1, authors of the reference argued that there exists innate bias in the method and proposed a non-iterative algorithm to reflect local distance information under a range of neighborhood sizes.

est.nearneighbor2(X, kmin = 2, kmax = max(3, round(ncol(X)/2)))

Arguments

X

an \((n\times p)\) matrix or data frame whose rows are observations.

kmin

minimum neighborhood size, larger than 1.

kmax

maximum neighborhood size, smaller than \(p\).

Value

a named list containing containing

estdim

estimated intrinsic dimension.

References

Verveer PJ, Duin RPW (1995). “An Evaluation of Intrinsic Dimensionality Estimators.” IEEE Transactions on Pattern Analysis and Machine Intelligence, 17(1), 81--86.

Author

Kisung You

Examples

# \donttest{
## create an example data with intrinsic dimension 2
X = cbind(aux.gensamples(dname="swiss"),aux.gensamples(dname="swiss"))

## acquire an estimate for intrinsic dimension
output = est.nearneighbor2(X)
sprintf("* est.nearneighbor2 : estimated dimension is %.2f.",output$estdim)
#> [1] "* est.nearneighbor2 : estimated dimension is 4.61."
# }