On the Liu and almost unbiased Liu estimators in the presence of multicollinearity with heteroscedastic or correlated errors

Journal Title: Surveys in Mathematics and its Applications - Year 2009, Vol 4, Issue 0

Abstract

This paper introduces a new biased estimator, namely, almost unbiased Liu estimator (AULE) of β for the multiple linear regression model with heteroscedastics and/or correlated errors and suffers from the problem of multicollinearity. The properties of the proposed estimator is discussed and the performance over the generalized least squares (GLS) estimator, ordinary ridge regression (ORR) estimator (Trenkler, 1984), and Liu estimator (LE) (Kaçiranlnar, 2003) in terms of matrix mean square error criterion are investigated. The optimal values of d for Liu and almost unbiased Liu estimators have been obtained. Finally, a simulation study has been conducted which indicated that under certain conditions on d, the proposed estimator performed well compared to GLS, ORR and LE estimators.

Authors and Affiliations

Mustafa Alheety, B. M. Golam Kibria

Keywords

Related Articles

Fixed points for multivalued contractions on a metric space

The purpose of this paper is to prove a fixed point theorem for multivalued operators and a fixed point theorem for multivalued weakly Picard operators in the terms of τ -distance.

Fixed point theorems for generalized weakly contractive mappings

In this paper several fixed point theorems for generalized weakly contractive mappings in a metric space setting are proved. The set of generalized weakly contractive mappings considered in this paper contains the family...

Homotopy analysis method for solving KdV equations

A scheme is developed for the numerical study of the Korteweg-de Vries (KdV) and the Korteweg-de Vries Burgers (KdVB) equations with initial conditions by a homotopy approach. Numerical solutions obtained by homotopy ana...

Fractional order differential inclusions on the half-line

We are concerned with the existence of bounded solutions of a boundary value problem on an unbounded domain for fractional order differential inclusions involving the Caputo fractional derivative. Our results are based o...

Download PDF file
  • EP ID EP107885
  • DOI -
  • Views 77
  • Downloads 0

How To Cite

Mustafa Alheety, B. M. Golam Kibria (2009). On the Liu and almost unbiased Liu estimators in the presence of multicollinearity with heteroscedastic or correlated errors. Surveys in Mathematics and its Applications, 4(0), 155-167. https://europub.co.uk/articles/-A-107885