Exploratory Study of the Application of Generalized Inverse to ILS Estimation of Overidentified Equations in Linear Models.

摘要

In this paper, the author proposes a procedure based on the use of the Moore-Penrose inverse of matrices for deriving unique Indirect Least Squares (ILS) estimates of the structural parameters in the overidentified case. The procedure makes use of all reduced form estimates in deriving the unique structural estimates. The estimator is shown to be consistent. The author derives the relationship between this Two-Stage Least Squares (2SLS) estimator and Instrumental-Variables (I.V.) estimators. Also derived is the asymptotic distribution of the proposed estimator. The results of sampling experiments are summarized.