摘要:

In recent years the Box-Jenkins method has become a popular technique for forecasting future behavior of a time series. Once adecruate computer packages are available for most purposes. un fortunately the problem of determining the appropriate forecast model has, for models of any complexity, been one of the major stumbling blocks to the user of this method.

ISSN:0361-0918    

DOI:10.1080/03610917808812057

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

摘要:

It is very pleasant to see what must often have been a major complication largely removed. (My choice of words indicate that I have not chosen p and q for an ARMA (p, q) process very often. ) As a bystander, I have two sorts of comments to make: (1) how we can make the procedure proposed by Gray et al even more perspicuous (specifically in picking which column of R’s or S’s to examine, and (2) what il might he in order to do now

ISSN:0361-0918    

DOI:10.1080/03610917808812058

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610917808812059

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610917808812060

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610917808812061

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610917808812062

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

摘要:

In this paper new ARMA models for the yearly sunspot numbers are obtained using the R- and S- arrays of Gray, Kelley, and Mclntire (1978). The models obtained are nonstationary and nearly nonstationary ARMA (8, 1) models. These new models are shown to be superior to the well-known ARMA (2, 0) model suggested by the Box-Jenkins approach.

ISSN:0361-0918    

DOI:10.1080/03610917808812063

出版商:Marcel Dekker, Inc.    

年代:1978

数据来源: Taylor

ISSN:0361-0918    

DOI:10.1080/03610917808812056

出版商:Marcel Dekker, Inc    

年代:1978

数据来源: Taylor