Two reduced-form approaches to the derivation of the maximum-likelihood estimators for simultaneous-equation systems (Q1058254)

This paper presents two new ways of deriving the maximum-likelihood estimators of subsystems of a simultaneous-equation model. The two derivations are based on an expression of the likelihood function in terms of the reduced-form parameters; and they give rise to two alternative versions of the estimating equations. These equations are compared with those that are derivable from the concentrated likelihood function of \textit{T. C. Koopmans} and \textit{W. C. Hood} [Studies econometric Method, 112-199 (1953; Zbl 0053.280)] and the equivalence of the three sets of estimating equations is demonstrated.