Maximum spacing estimation for multivariate observations under a general class of information-type measures

Sammanfattning

This article considers the maximum spacing (MSP) method for multivariate observations, nearest neighbour balls are used as a multidimensional analogue to univariate spacings. Compared to the previous studies, a broader class of MSP estimators corresponding to different information-type measures is studied. The studied class of estimators includes also the estimator corresponding to the Kullback-Leibler information measure obtained with the logarithmic function. Consistency of the MSP estimators is proved when the assigned model class is correct, that is the true density belongs to the assigned class. The behaviour of the MSP estimator under different divergence measures is studied and the advantage of using MSP estimators corresponding to different information measures in the context of model validation is illustrated in simulation examples.

Nyckelord

Consistency; Divergence measures; Maximum spacing estimation; Model validation Nearest neighbour; Nearest neighbour balls

Publicerad i

Journal of Multivariate Analysis
2025, volym: 208, artikelnummer: 105433
Utgivare: ELSEVIER INC

SLU författare

• Ranneby, Bo

  • Institutionen för skogsekonomi, Sveriges lantbruksuniversitet
    

UKÄ forskningsämne

Sannolikhetsteori och statistik

Publikationens identifierare

• DOI: https://doi.org/10.1016/j.jmva.2025.105433

Permanent länk till denna sida (URI)

https://res.slu.se/id/publ/141461