idpettis

Returns the estimate of intrinsic dimensionality using the algorithm of Pettis, Bailey, Jain, and Dubes, 1979. It is based on the nearest-neighbor distances of a data set.

Input Arguments:

• ydist is a vector that contains the nearest neighbor distances as obtained by the Statistics Toolbox function pdist.
• n is the sample size.

Output Arguments:

idhat is the estimate of the intrinsic dimensionality of the data.

Synopsis

idhat = idpettis(ydist,n)

References:

Pettis, K. W., T. A. Bailey, A. K. Jain, and R. C. Dubes. 1979. “An intrinsic dimensionality estimator from near-neighbor information,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 1:25-37.