You are not currently logged in.
Access JSTOR through your library or other institution:
Discrete Multivariate Distributions and Generalized Log-Concavity
R. B. Bapat
Sankhyā: The Indian Journal of Statistics, Series A (1961-2002)
Vol. 50, No. 1 (Feb., 1988), pp. 98-110
Published by: Indian Statistical Institute
Stable URL: http://www.jstor.org/stable/25050683
Page Count: 13
You can always find the topics here!Topics: Eigenvalues, Matrices, Binomials, Integers, Random variables, Mathematical inequalities, Mathematical sequences, Numbers, Distribution functions, Mathematics
Were these topics helpful?See something inaccurate? Let us know!
Select the topics that are inaccurate.
Preview not available
For discrete multivariate distributions we introduce a notion called generalized log-concavity and then show that several standard distributions satisfy that property. The multiparameter multinomial density is shown to be generalized log-convave and the proof depends on results from the theory of permanents. The multiparameter negative multinomial density is expressed in terms of permanents and it is shown that the multiparameter negative binomial density is log-concave. The Alexandroff inequality for permanents is used to show that certain sequences associated with the multiparameter multinomial and with order statistics for nonidentically distributed variables are log-concave.
Sankhyā: The Indian Journal of Statistics, Series A (1961-2002) © 1988 Indian Statistical Institute