Unlike the usual stochastic order, total positivity order is closed under conditioning. Here we provide a general formulation of the preservation properties of the order under conditioning; we study certain properties of the order including translation properties and the implications of having equality in the inequality defining the order. Specializing to the multivariate normal distribution, the study of total positivity order leads to new cones defined in terms of covariance M-matrices related to positive dependence, whose properties we study.
Total positivity order and the normal distribution / Rinott, Joseph; Scarsini, Marco. - In: JOURNAL OF MULTIVARIATE ANALYSIS. - ISSN 0047-259X. - 97:(2006), pp. 1251-1261. [10.1016/j.jmva.2005.07.008]
Total positivity order and the normal distribution
RINOTT, JOSEPH;SCARSINI, MARCO
2006
Abstract
Unlike the usual stochastic order, total positivity order is closed under conditioning. Here we provide a general formulation of the preservation properties of the order under conditioning; we study certain properties of the order including translation properties and the implications of having equality in the inequality defining the order. Specializing to the multivariate normal distribution, the study of total positivity order leads to new cones defined in terms of covariance M-matrices related to positive dependence, whose properties we study.File | Dimensione | Formato | |
---|---|---|---|
JMVA2006RS.pdf
Solo gestori archivio
Tipologia:
Documento in Post-print
Licenza:
DRM (Digital rights management) non definiti
Dimensione
165.74 kB
Formato
Adobe PDF
|
165.74 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.