Knowledge that Transforms

This paper specifies and estimates a four-equation disequilibrium model of the consumption goods market in a centrally planned economy (CPE).The data are from Poland for the period 1955-1980, but the analysis is more general and will be applied to other CPEs as soon as the appropriate data sets are complete.This work is based on previous papers of Portes and Winter (P-W) and Charemza and Quandt (C-Q).P-W applied to each of four CPEs a discrete-switching disequilibrium model with a household demand equation for consumption goods, a planners' supply equation, and a "mm" condition stating that the observed quantity transacted is the lesser of the quantities demanded and supplied.C-Q considered how an equation for the adjustment of planned quantities could be integrated into a CPE model with fixed prices and without the usual price adjustment equation.They made plan formation endogenous and permitted the resulting plan variables to enter the equations determining demand and supply.This paper implements the C-Q proposal in the P-W context.It uses a unique new data set of time series for plans for the major macroeconomic variables in Poland and other CPEs.The overall framework is applicable to any large organization which plans economic variables.

This paper characterizes all utility derived demand systmes having Engel curves that are linear in both income and an arbitrary function of income. This class encompasses virtually all utility derived demand systems that have been estimated in the past using ag gregate data with explicit treatment of the problem of aggregation ac ross individuals. It includes extensions of the PIGLOG and PIGL class es that have similar properties to these classes, but allow for more general Engel curve shapes. This paper extends W. M. Gorman's study o f these forms primarily by characterizing systems of rank two. The ap plication of the characterized systems to problems of nesting, separa bility, flexibility, Engel curve analysis, estimation, and aggregatio n are briefly discussed.

Previous analyses have shown that if a point is to be a core of a majority rule voting game in Euclidean space, when preferences are smooth, then the utility gradients at the point must satisfy certain restrictive symmetry conditions. In this paper, these results are generalized to the case of an arbitrary voting rule, and necessary and sufficient conditions, expressed in terms of the utility gradients of "pivotal" coalitions, are obtained.

On montre que l'estimateur de Manski-McFadden (1981) est asymptotiquement plus efficace que l'estimateur de Manski-Lerman (1977)

The authors discuss the unity between the two standard approaches to noncooperative solution concepts for games. The decision-theoretic approach starts from the assumption that the rationality of the players is common knowledge. This leads to the notion of correlated rationalizability. It is shown that correlated rationalizability is e quivalent to a posteriori equilibrium-a refinement of subjective corr elated equilibrium. Hence a decision-theoretic justification for the equilibrium approach to game theory is provided. An analogous equival ence result is proved between independent rationalizability, which is the appropriate concept if each player believes that the others act independently, and conditionally independent a posteriori equilibrium.

In a multiple-regression model the residual variance is an unknown function of the explanatory variables, and estimated by nearest-neighbor nonparametric regression. The resulting weighted least-squares estimator of the regression coefficients is shown to be adaptive, in the sense of having the same asymptotic distribution, to first order, as estimators based on knowledge of the actual variance function or a finite parameterization of it. A similar result was established by R. J. Carrol l (1982) using kernel estimation and under substantially more restrictive conditions on the data generating process than ours. Extensions to various other models seem to be possible.

The analogy principle proposes that population parameters be estimated by sample statistics which make known properties of the population hold as closely as possible in the sample. Applications of the analogy principle are ubiquitous. Nevertheless, estimation theory has not been studied from a consistent analog perspective. This paper makes a start.

Buyer's preferences over auctions depend on their measure of absolute risk aversion. If it is constant and they have independent private values, they are indifferent between a first-price auction (FPA), a second-price auction (SPA), and a first-price auction in which the number of bidders is revealed before bids are taken (FPA- R). If they have decreasing absolute risk aversion, they prefer the SPA to the FPA-R to the FPA. Their preference for the SPA is diminished to the extent that their values are affiliated. Affiliation also causes them to prefer the number of bidders to be revealed, whereas the seller then prefers to keep it secret.

When a decision rule is implemented using a Bayesian incentive compatible mechanism in which the messages are publically obser vable, the players' information is augmented by their observation of each others' strategies. In this paper the authors study the set of Bayesian implementable decision rules which have the further property that the information c onveyed in the process of thier implementation does not invalidate the optimality of the players' strategies. Such rules are called posterior implementable. The authors concentrate on a two-person problem with two possible decisions and, for this problem, they obtain a complete characterization of the set of posterior implementable decision rules.