eduzhai > Physical Sciences > Physics Sciences >

Pseudodistance Methods Using Simultaneously Sample Observations and Nearest Neighbour Distance Observations for Continuous Multivariate Models

  • Save

... pages left unread,continue reading

Document pages: 13 pages

Abstract: Using the fact that a multivariate random sample of n observations alsogenerates n nearest neighbour distance (NND) univariate observations and fromthese NND observations, a set of n auxiliary observations can be obtained andwith these auxiliary observations when combined with the original multivariateobservations of the random sample, aclass of pseudodistance Dh is allowed to be used and inference methods can be developed using this class ofpseudodistances. The Dh estimators obtainedfrom this class can achieve high efficiencies and have robustness properties.Model testing also can be handled in a unified way by means of goodness-of-fittests statistics derived from this class which have an asymptotic normaldistribution. These properties make the developed inference methods relativelysimple to implement and appear to be suitable for analyzing multivariate datawhich are often encountered in applications.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×