Consistent Estimation of Regression Coefficients in Replicated Data with Non-Normal Measurement Errors

Aman Ullah

and  Debasri Mukherjee

In this paper we consider a weighted harmonic mean of two inconsistent estimators to propose a new estimator of the coefficient of a linear regression model with measurement errors. The proposed estimator is simple and it does not depend on any unknown quantity. The approximate bias and MSE of the estimator are derived. Further, an empirical application is also presented.

Key Words: Measurement Error; Replicated data, Nonnormal erros;
JEL Classification Numbers: C1.