Restrictions on variables
- 30 April 1982
- journal article
- Published by Elsevier BV in Journal of Econometrics
- Vol. 18 (3), 369-393
- https://doi.org/10.1016/0304-4076(82)90089-6
Abstract
Accounting identities impose exact restrictions on the endogenous variables of econometric models. Such restrictions are usually met by choosing a closing entry or by building an allocation model. Selecting a closing entry may be difficult or arbitrary, while allocation models admit little flexibility in the choice of explanatory variables and lagged adjustment schemes. This paper studies a third solution which in a sense lies inbetween: freely chosen equations for all variables are adjusted additively such that the restrictions will hold. The adjustment involves some new parameters which can be estimated simultaneously with the original parameters using Maximum Likelihood techniques. An application is provided for a financial model of the Dutch private sector. Our approach here proves superior to any choice of closing entry in the system.Keywords
This publication has 10 references indexed in Scilit:
- A note on the compatibility of a system of difference equations and a time-independent linear equationEconomics Letters, 1978
- Problèmes d'estimation d'un modèle économétrique vectorialiséCahiers du Séminaire d'Économétrie, 1977
- Estimation and Hypothesis Testing in Singular Equation Systems with Autoregressive DisturbancesEconometrica, 1975
- Consistent Sum-Constrained ModelsJournal of the American Statistical Association, 1975
- The Differentiation of Pseudo-Inverses and Nonlinear Least Squares Problems Whose Variables SeparateSIAM Journal on Numerical Analysis, 1973
- Maximum Likelihood Estimation of the Linear Expenditure SystemJournal of the American Statistical Association, 1971
- Some Theorems on Matrix Differentiation with Special Reference to Kronecker Matrix ProductsJournal of the American Statistical Association, 1969
- On matrix procedures for optimizing differentiable scalar functions of matricesStatistica Neerlandica, 1967
- An Efficient Method of Estimating Seemingly Unrelated Regressions and Tests for Aggregation BiasJournal of the American Statistical Association, 1962
- Maximum-Likelihood Estimation of Parameters Subject to RestraintsThe Annals of Mathematical Statistics, 1958