Testing the restrictions of the Almon lag technique

L. G. Godfrey, D. S. Poskitt

Research output: Contribution to journalArticleResearchpeer-review

20 Citations (Scopus)


This article considers testing the restrictions imposed by Almon’s approach to estimation of distributed lag relationships. The conventional likelihood ratio test procedure requires that the regression equation be estimated in unrestricted form and then subject to the Almon restrictions. An alternative test is suggested which has the advantage of requiring only that the unrestricted version be estimated. The problem of selecting the order of the polynomial is discussed and a numerical example is presented.

Original languageEnglish
Pages (from-to)105-108
Number of pages4
JournalJournal of the American Statistical Association
Issue number349
Publication statusPublished - 1 Jan 1975
Externally publishedYes

Cite this