Academic Journal

1Generalized Diagonal Band Copulae with Two-Sided Generating Densities

Bibliographic Details
Title: 1Generalized Diagonal Band Copulae with Two-Sided Generating Densities
Authors: Samuel Kotz, J. R. Van Dorp, A J. R. Van Dorp
Contributors: The Pennsylvania State University CiteSeerX Archives
Superior Title: http://business.gwu.edu/decisionsciences/i2sds/pdf/TR-2009-10.pdf.
Publication Year: 2009
Collection: CiteSeerX
Description: Maximum entropy copulae introduced by Bedford and Meeuwissen (1997) provide normative experts the option of making minimally informative assumptions given a degree of dependence constraint between two random variables. Unfortunately, their distributions functions are not available in a closed form and application requires the use of numerical methods. In this paper we shall study a sub-family of generalized diagonal band (GDB) copulae, separately introduced by Ferguson (1995) and Bojarski (2001). Specifically, bivariate copulae density support shall cover the complete unit square when a full support generating density is selected from the two-sided (TS) framework of distributions introduced by Van Dorp and Kotz (2003). Even with unit square support, GDB copulae with TS generating densities allow for a complete rank correlation coverage by taking advantage of the TS framework's flexibility. GDB copulae distribution functions and properties shall be expressed in terms of the TS generating density. Instances shall be presented with closed form expressions utilizing only elementary functions. A straightforward expert judgment elicitation procedure for the GDB copula dependence parameter is suggested. Operationally, for rank correlations ranging from to, GDB copulae with TS Slope generating densities !Þ %!Þ% closely approximate the minimal information measure of the maximum entropy copulae. For absolute rank correlation values larger than, a TS power generating density for GDB copulae!Þ% performs very well in this regard. For demonstration purposes the sub-family of GDB copulae discussed herein is illustrated in a decision analysis example.
Document Type: text
File Description: application/pdf
Language: English
Relation: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.624.8251; http://business.gwu.edu/decisionsciences/i2sds/pdf/TR-2009-10.pdf
Availability: http://business.gwu.edu/decisionsciences/i2sds/pdf/TR-2009-10.pdf
Rights: Metadata may be used without restrictions as long as the oai identifier remains attached to it.
Accession Number: edsbas.A9658A0C
Database: BASE
Description
Description not available.