[ Login ]

User Publications  

Items are presented in alphabetical order by author's name. Deutsch, R., Cherner, M., Grant, I. [June] 2006. Significance testing of a cluster of multivariate binary variables: comparison of the tripartite T index to three common similarity measures. Stat. Meth. Medic. Res. 15(3): 285-299. Comments: This paper is a good evaluation of the tripartite similarity metric that brings out a factor important in medicine as opposed to taxonomy. When something is absent from sets of symptoms, it is called "health," and the number of healthy patients in populations may be important to a medical study. Hence the Tripartite metric [like all the metrics classified as "Similarity-dissimilarity coefficients" by L. C. Hayek (1994)] used with "healthy" as equivalent to "absent" can be problematic because "joint absence" is not addressed by the metrics. In at least some cases, if absence of a set element is important, an element can be introduced such as "lacking all the symptoms" or "body not striped in black." The paper tests the T-similarity metric against others and finds that the Dice metric produces almost identically linear output date when compared with ouput of the T-similarity metric for real medical data sets. The first author has expressed to me that, in her view, relative size of the sets being compared need not be taken into account in a similarity metric. She is not concerned with aliasing in her applications. The T similarity metric is criticized for its computational complexity (motivated by the desire to eliminate aliasing) and for the lack of an obvious interpretation for the values it produces as output. -- R. E. T. Kong, A. and A. Estrada-Torres. 2002. El Género Lactarius en Tlaxcala, México In: Guzman, G. and D. Mata. Nanacatepec. Estudios sobre los hongos latinoamericanos. Resúm. Congr. Latinoamer. Micol. 9: 227. [abstract] Pécheux, L. le M. 2002. Modèle d'estimation de la biodiversité au niveau du paysage: LandBioDiv. Une possibilité pour l'étude de la relation hétérogénéité spatiale/diversité végétale (thèse). Inst. Medit. Écol. Paléoécol. [abstract: imep-cnrs.com/pages/imep1.htm] Schils, T. 2006. The tripartite biogeographical index: a new tool for quantifying spatio-temporal differences in distribution patterns. J. Biogeogr. 33: 560-572. Schils, T. and E. Coppejans. 2003. Phytogeography of upwelling areas in the Arabian Sea. J. Biogeogr. 30(9): 1339-1356. [abstract: blackwellpublishing.com/journals/jbi/, requires limited permission for cookies] Tulloss, R. E. 2002. Amanita -- studies on distribution of a taxa-rich genus in the Americas In: Guzman, G. and G. Mata, eds., Estudios sobre los hongos latinoamericanos. Resúm. Congr. Latinoamer. Micol. 9: 52. [abstract] Tulloss, R. E. 2005. Amanita -- distribution in the Americas with comparison to eastern and southern Asia and notes on spore character variation with latitude and ecology. Mycotaxon 93: 189-231. Tulloss, R. E. and D. C. Tulloss. 2004. An on-line tripartite similarity calculator. Inoculum 55(1): 1-4. [http://www.msafungi.org/55(1).pdf]

© 2003, Amanita-Bear Consulting.
- Last modified: 06/15/2004