Computational Aspects of the Nearest Neighbor Statistics

doi:https://doi.org/10.1007/978-3-662-26811-7_15

Book Chapter10.1007/978-3-662-26811-7_15

Computational Aspects of the Nearest Neighbor Statistics

Elzbieta Trybus,G. Trybuś-1992-01-01-Computational Statistics

0

TL;DRAbstract

The distributions of the nearest neighbor statistics and its parameters are calculated in the case when points are generated from the multivariate uniform and multivariate standard normal distribution. A recursive function for the expected value of the k-th nearest neighbor is derived for the asymptotic distributions. Monte Carlo method is used to assess the presented approach.

Chat with Paper

AI Agents for this Paper

The distributions of the nearest neighbor statistics and its parameters are calculated in the case when points are generated from the multivariate uniform and multivariate standard normal distribution. A recursive function for the expected value of the k-th nearest neighbor is derived for the asymptotic distributions. Monte Carlo method is used to assess the presented approach.

Keywords

k-nearest neighbors algorithmMultivariate statisticsMonte Carlo methodStatisticsNearest neighbourMathematicsStatistical physicsComputer science

Chat

Click to start Chat