Paul A. Ruud, Sufficient Conditions for the Consistency of Maximum Likelihood Estimation Despite Misspecification of Distribution in Multinomial Discrete Choice Models, Econometrica, Vol. 51, No. 1 (Jan., 1983), pp. 225-228
[This paper presents an expected utility theory for state-dependent preferences. It proposes axioms that permit the joint derivation of subjective probabilities and utilities when the decision maker's preferences are not independent of the prevailing state of nature. In addition to the usual von Neumann-Morgenstern axioms, these axioms also include the requirement that the decision-maker's actual preferences are consistent with his preferences contingent on an hypothetical probability distribution over the states of nature. Two versions of the consistency axiom are introduced and their significance in the context of Bayesian decision theory is discussed.]
[Using smooth profit functions to characterize production possibilities, we extend the concepts of regularity and fixed point index to economies with very general technologies, involving both constant and decreasing returns. To prove the genericity of regular economies we rely on an approach taken by Mas-Colell that utilizes the topological concept of transversality. We also generalize the index theorem given by Kehoe. Our results shed new light on the question of when an economy has a unique equilibrium.]
[This paper characterizes a market economy with infinitely long-lived consumers, and value-maximizing firms which face costs of adjustment for capital. The temporary equilibrium of this economy is similar to the short-run equilibrium of standard macroeconomic models. Consumption is a function of wealth, investment is related to the value of firms; equilibrium between aggregate demand and aggregate supply is achieved by the endogenous adjustment of the sequence of current and future interest rates. The dynamic behavior of output, consumption, and investment in this economy is the same as in an optimal growth model with adjustment costs. The paper shows this equivalence and then uses it, together with the equivalence of taxes to technological shocks, to study the dynamic effects of fiscal policy.]
[This paper develops a framework of analysis for studying the informational properties of a certain class of "parametric" resource allocation processes. It is shown that the Taylor process (related to certain ideas for planning in the so-called socialist economies as put forward by Taylor [126] is informationally efficient in the sense that any informationally decentralized resource allocation process which has similar (static) properties (Pareto optimality) must use a message space which is dimensionally at least as large as that of the Taylor process. We also show that in general greater informational decentralization can be achieved through parametric than through "nonparametric" processes.]
THIS PAPER IS INTENDED to stress the distinction between the conditions for lack of identification in models linear with respect to the variables but nonlinear in the parameters in the sense originally defined by Fisher [2], and the less numerous set of conditions required for first order lack of identification. The latter set of conditions involve only the first derivatives of the coefficients as functions of the parameters. It is argued that if the model suffers from first order lack of identification, it will generally be the case that the usual estimators are consistent, although not asymptotically normally distributed. In a leading special case the asymptotic distribution is discussed, and the simulation of a simple model illustrates the extent to which this asymptotic distribution approximates the actual finite sample distribution.
Edlefsen [3] has shown that a phenomenon of great similarity to the strong LeChatelier principle can be established when altering the feasible set of an optimizing agent by suitably replacing existing constraints rather than adding new ones. Here it was demonstrated that essentially the same phenomenon occurs when altering the objective function in a systematic manner rather than the.feasible set. The result obtained is general enough to allow a replication of Edlefsen's analysis of a household choosing between quantity and quality when facing hedonic prices, and served here furthermore in an application to the theory of the firm where it was demonstrated that a systematic relationship exists between the intensity of the reactions of a producer to changes in a conjugate parameter and the conditions prevailing in his markets.
Particularly under the assumption of rational expectations, a model may have serially correlated errors and those errors may be uncorrelated with contemporaneous and lagged values of a predetermined instrument, yet the instruments may not be strictly exogenous. This paper proposes a method for transforming such a model to one without serial correlation, while keeping the instrument predetermined. Standard theory of instrumental variables estimation then applies. Furthermore, it turns out that for transformations of the class proposed, asymptotic distribution theory is the same whether the serial correlation properties of the errors are known a priori or estimated. As the number of lagged values of the predetermined variables used as instruments increases, the asymptotic variance of the standard instrumental variables estimator applied to the transformed model approaches that of the optimal estimator proposed by Hansen and Sargent [8]. IN A NUMBER of recently developed macroeconomic models behavioral equations arise in which error terms can be asserted on the basis of economic arguments to be uncorrelated with some set of instrumental variables at a certain set of dates, but not to be uncorrelated with the instruments at all dates. Examples of such
[Strategic considerations may induce a resource importing country to invent a substitute earlier than it intends to put it to use. There are also circumstances in which it would wish to delay an invention date even if it could obtain it at an earlier date at no extra cost. Similar paradoxical results obtain if resource cartels behave strategically. Setting prices high may be a way of deterring invention. If those engaged in R & D are not resource users, and the cartel has access to similar R & D technology, it will pre-empt rivals. This may not be the case if resource users can also engage in R & D.]