Abstract. This paper introduces a new Estimator for multicollinearly matrix data and autocorrelated errors. We purpose two Stages Ridge Estimator(TR) for the multiple linear regression model, which suffers from both problems autocorrelation (AR(1)) and multicollinearity. After adjusting this with the ordinary ridge regression estimator (ORR), we use a mixed method to apply the two stages least squares procedure (TS). We also derive some statistical properties of this biased estimator and the paper is achieved by an application example.

Received: October 23, 2012

AMS Subject Classification: 62J07, 62J05

Key Words and Phrases: two stages estimator, multicollinearity, singular matrices, relaxation method, autocorrelated errors, general linear models

Download paper from here.