Full metadata record
DC FieldValueLanguage
dc.contributor.authorChen, ZYen_US
dc.contributor.authorChou, YCen_US
dc.date.accessioned2014-12-08T15:45:11Z-
dc.date.available2014-12-08T15:45:11Z-
dc.date.issued2000-06-28en_US
dc.identifier.issn0167-9473en_US
dc.identifier.urihttp://hdl.handle.net/11536/30454-
dc.description.abstractBased on the recasting techniques of Rust and Volt (1990, J. Amer. Statist. Assoc. 85, 572-578), an S-system form of the noncentral beta distribution is extended from that of the noncentral F distribution and the other one is newly derived. The computing methods of this distribution have received much attention during the last decade. Its cumulative probabilities, densities, quantiles and related distributional values can be calculated in one S-system form. We demonstrate the new computational results using the S-system numerical solver ESSYNS. Consistent results are obtained from these two S-system forms under various situations. In addition, we compare the performance with an ad hoc computing method by evaluating the cumulative probabilities and densities jointly. The S-system formulation provides significant numerical advantages over its original form. Further properties are also discussed (C) 2000 Elsevier Science B.V. All rights reserved.en_US
dc.language.isoen_USen_US
dc.subjectstatistical computingen_US
dc.subjectdifferential equationen_US
dc.subjectrecastingen_US
dc.subjectpower lawen_US
dc.subjectTaylor seriesen_US
dc.titleComputing the noncentral beta distribution with S-systemen_US
dc.typeArticleen_US
dc.identifier.journalCOMPUTATIONAL STATISTICS & DATA ANALYSISen_US
dc.citation.volume33en_US
dc.citation.issue4en_US
dc.citation.spage343en_US
dc.citation.epage360en_US
dc.contributor.department管理科學系zh_TW
dc.contributor.departmentDepartment of Management Scienceen_US
dc.identifier.wosnumberWOS:000087784300001-
dc.citation.woscount5-
Appears in Collections:Articles


Files in This Item:

  1. 000087784300001.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.