On the specification of error structure in the linear regression model

Kuldeep Kumar, Mukesh Srivastava

Research output: Contribution to journalArticleResearchpeer-review

Abstract

The assumption regarding the independence of errors in the linear regression model may not be valid quite often. In this paper, we have discussed a new method based on the theory of Pade approximation for the specification of the error structure which may follow an ARMA (p, q) model. We have also discussed the consequences of the error and reviewed the existing methods for the specification of the error. Using simulation studies we have shown that this method works well for the specification of the error structure.
Original languageEnglish
Pages (from-to)105-133
Number of pages29
JournalAdvances and applications in statistics
Volume3
Issue number2
Publication statusPublished - 2003

Fingerprint

Dive into the research topics of 'On the specification of error structure in the linear regression model'. Together they form a unique fingerprint.

Cite this