ESTIMATION IN MULTIFACTOR POLYNOMIAL REGRESSION UNDER NON-NORMALITY


Tiku M. L., Akkaya A.

PAKISTAN JOURNAL OF STATISTICS, vol.26, no.1, pp.49-68, 2010 (SCI-Expanded) identifier

  • Publication Type: Article / Article
  • Volume: 26 Issue: 1
  • Publication Date: 2010
  • Journal Name: PAKISTAN JOURNAL OF STATISTICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.49-68
  • Keywords: Non-normality, Maximum likelihood, Modified maximum likelihood, Robustness, Polynomial regression, Outliers, Inliers, MAXIMUM-LIKELIHOOD, ROBUST ESTIMATION, DISTRIBUTIONS, PARAMETERS, MODEL, IDENTIFICATION, LOCATION
  • Middle East Technical University Affiliated: Yes

Abstract

Modified maximum likelihood estimators of the parameters in a second order polynomial regression model are derived. They are shown to be considerably more efficient and robust than the commonly used least squares estimators. Real life examples are given.