We present a general theory of consistent estimation for possibly misspecified parametric models based on recent results of Domowitz and White. This theory extends the unification of Burguete, Gallant, and Souza by allowing for heterogeneous, time-dependent data and dynamic models. The theory is applied to yield consistency results for quasi-maximum-likelihood and method of moments estimators. Of particular interest is a new generalized rank condition for identifiability.

Let X0 be a square matrix (complex or otherwise) and u0 a (normalized) eigenvector associated with an eigenvalue λo of X0, so that the triple (X0, u0, λ0) satisfies the equations Xu = λu, . We investigate the conditions under which unique differentiable functions λ(X) and u(X) exist in a neighborhood of X0 satisfying λ(X0) = λO, u(X0) = u0, Xu = λu, and . We obtain the first and second derivatives of λ(X) and the first derivative of u(X). Two alternative expressions for the first derivative of λ(X) are also presented.

Methods of maximum-likelihood search (eg., Hildreth-Lu) for single equations models with autocorrelated disturbances are not easily extended to the multi-equation systems encountered in econometric studies of consumer demand. The obstacle is the large dimension of the space of independent serial correlation coefficients to be partitioned and searched. This paper treats stochastic disturbance vectors in demand/expenditure systems as functions of random disturbances in consumer utility functions. It specifies marginal utilities to be products of a systematic function and a nonnegative random disturbance. Under this specification, three hypotheses that readily come to mind in the context of consumer theory drastically reduce the number of independent elements of serial covariance matrices. This in turn makes practicable the extension of maximum-likelihood search methods in the estimation of parameters of large demand systems.

This paper reconsiders King's [12] locally optimal test procedure for first-order moving average disturbances in the linear regression model. It recommends two tests, one for problems involving positively correlated disturbances and one for negatively correlated disturbances. Both tests are most powerful invariant at a point in the alternative hypothesis parameter space that is determined by a function involving the sample size and the number of regressors. Selected bounds for the tests' significance points are tabulated and an empirical comparison of powers demonstrates the overall superiority of the new test for positively correlated moving average disturbances.

In this paper we consider the situation in which ordinary least squares (OLS) is used to estimate an ARMA (1,1) model with one exogenous variable. Applying Edgeworth expansion techniques, we examine the misspecification errors and the approximate distributions of the OLS estimator. Extensive numerical studies were performed and selected results are shown graphically. In addition, a technical device is developed to calculate the Edgeworth coefficients.

This paper is concerned with Cornish–Fisher corrections of some instrumental variables test statistics. The tests based on the corrected statistics have size with error of a smaller order of magnitude than the original tests. Symmetric Edgeworth-corrected confidence regions are also defined for the structural parameters. All these corrections are given as analytic formulas that require only limited information, so their implementation is a relatively easy task.