Show simple item record

dc.contributor.authorZhen, Xiaoa
dc.date.accessioned2014-03-04T16:20:06Z
dc.date.available2014-03-04T16:20:06Z
dc.date.issued2008-08
dc.identifier.otherzhen_xiaoa_200808_phd
dc.identifier.urihttp://purl.galileo.usg.edu/uga_etd/zhen_xiaoa_200808_phd
dc.identifier.urihttp://hdl.handle.net/10724/25103
dc.description.abstractStandard time series models typically assume that the data are continuous. If the available data consist of counts of observations in a nite number of categories, the usual autoregressive moving average (ARMA) models cannot be applied to t the count data. For each time t, the count vector N(t) = (N1(t);N2(t);
dc.languageeng
dc.publisheruga
dc.rightspublic
dc.subjectBinary Date
dc.subjectPartial Likelihood
dc.subjectCategorical Time Series
dc.subjectMaximum Likelihood
dc.subjectState Space Models
dc.subjectDependent Contingency Tables
dc.titleCategorical time series
dc.typeDissertation
dc.description.degreePhD
dc.description.departmentStatistics
dc.description.majorStatistics
dc.description.advisorIshwar Basawa
dc.description.committeeIshwar Basawa
dc.description.committeeLynne Seymour
dc.description.committeeJaxk Reeves
dc.description.committeeDaniel Hall
dc.description.committeeGauri Datta


Files in this item

FilesSizeFormatView

There are no files associated with this item.

This item appears in the following Collection(s)

Show simple item record