Please use this identifier to cite or link to this item: https://scholarbank.nus.edu.sg/handle/10635/105436
DC FieldValue
dc.titleTime series models with asymmetric innovations
dc.contributor.authorTiku, M.L.
dc.contributor.authorWong, W.-K.
dc.contributor.authorBian, G.
dc.date.accessioned2014-10-28T05:16:05Z
dc.date.available2014-10-28T05:16:05Z
dc.date.issued1999
dc.identifier.citationTiku, M.L.,Wong, W.-K.,Bian, G. (1999). Time series models with asymmetric innovations. Communications in Statistics - Theory and Methods 28 (6) : 1331-1360. ScholarBank@NUS Repository.
dc.identifier.issn03610926
dc.identifier.urihttp://scholarbank.nus.edu.sg/handle/10635/105436
dc.description.abstractWe consider AR(q) models in time series with asymmetric innovations represented by two families of distributions: (i) gamma with support IR : (0, ∞), and (ii) generalized logistic with support IR : (-∞, ∞). Since the ML (maximum likelihood) estimators are intractable, we derive the MML (modified maximum likelihood) estimators of the parameters and show that they are remarkably efficient besides being easy to compute. We investigate the efficiency properties of the classical LS (least squares) estimators. Their efficiencies relative to the proposed MML estimators are very low. Copyright © 1999 by Marcel Dekker, Inc.
dc.sourceScopus
dc.subjectGamma distribution
dc.subjectGeneralized logistic
dc.subjectHypothesis testing
dc.subjectModified likelihood
dc.subjectNonnormality
dc.subjectPower function
dc.subjectRobustness
dc.subjectTime series
dc.typeArticle
dc.contributor.departmentSTATISTICS & APPLIED PROBABILITY
dc.description.sourcetitleCommunications in Statistics - Theory and Methods
dc.description.volume28
dc.description.issue6
dc.description.page1331-1360
dc.description.codenCSTMD
dc.identifier.isiutNOT_IN_WOS
Appears in Collections:Staff Publications

Show simple item record
Files in This Item:
There are no files associated with this item.

Page view(s)

112
checked on Jan 20, 2022

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.