do.made first aims at finding local dimesion estimates using nearest neighbor techniques based on the first-order approximation of the probability mass function and then combines them to get a single global estimate. Due to the rate of convergence of such estimate to be independent of assumed dimensionality, authors claim this method to be manifold-adaptive.

est.made(
X,
k = round(sqrt(ncol(X))),
maxdim = min(ncol(X), 15),
combine = c("mean", "median", "vote")
)

## Arguments

X an $$(n\times p)$$ matrix or data frame whose rows are observations. size of neighborhood for analysis. maximum possible dimension allowed for the algorithm to investigate. method to aggregate local estimates for a single global estimate.

## Value

a named list containing containing

estdim

estimated global intrinsic dimension.

estloc

a length-$$n$$ vector estimated dimension at each point.

## References

Farahmand AM, Szepesvári C, Audibert J (2007). “Manifold-Adaptive Dimension Estimation.” In ICML, volume 227 of ACM International Conference Proceeding Series, 265--272.

Kisung You

## Examples

# \donttest{
## create a data set of intrinsic dimension 2.
X = aux.gensamples(dname="swiss")

## compare effect of 3 combining scheme
line1 = paste0("* est.made : 'mean'   estiamte is ",round(out1$estdim,2)) line2 = paste0("* est.made : 'median' estiamte is ",round(out2$estdim,2))