Adjusted Linear Estimator: An Application to Farm Management Data

For errors in variables problem, an Adjusted Linear Estimator (ALE) of parameter vector of linear model is proposed when the true variables and errors follow different distributions under the assumption that linearity is maintained, at least approximately, in the observed variables. The main objective of this paper is to see how this ALE is useful as an estimator of the parameter vector of the linear model in non-normal case. Extensive Monte Cario Study is made using Extended Ridge Method (ERM).

From the point of view of Bias and Mean Square Error (MSE), the performance of ALE is found to be better than OLE. This paper also examines the application of an Adjusted Linear Estimator to the analysis of Farm Management Data. The problem of estimation of Errors in variables linear model is discussed for a Cobb-Douglas Production Function. The Adjusted Linear Estimates of input elasticities are computed and then compared with the corresponding Oridinary Least Squares estimates. It is also proved that the estimated variance of Adjusted Linear Estimator is smaller than that of the Oridinary Least Squares Estimator.