Knowledge that Transforms

Robust estimation aims at developing point estimators that are not highly sensitive to errors in data. However, the population parameters of interest are not identified under the assumptions of robust estimation, so the rationale for point estimation is not apparent. This paper shows that, under error models used in robust estimation, unidentified population parameters can often be bounded. The bounds provide information that is not available in robust estimation. For example, it is possible to bound the population mean under contaminated sampling. It is argued that estimating the bounds is more natural than attempting point estimation of unidentified parameters. Copyright 1995 by The Econometric Society.

"In this paper, we formulate a statistical model of dynamic intrafamily investment behavior incorporating endowment heterogeneity and heritability. We use the model's estimates to evaluate alternative estimation procedures that have exploited family and kinship data to obtain estimates of the determinants of human capital." The sequential decision-making framework developed is applied to data on birthweight and gestation of children born to mothers surveyed in the U.S. National Survey of Labor Market Experience youth cohort. "The empirical results imply that the least restrictive statistical formulation, consistent with dynamic behavior and heterogeneity among siblings, fits the data best. All of the estimation procedures that control for a family-specific endowment indicate, however, that the biological effect of having a birth at younger ages is to marginally increase birthweight and to increase fetal growth."

This paper considers the theory of market versus optimal product diversity in the light of two recent advances in oligopoly theory. The first is the development of discrete choice models to describe heterogeneous consumer tastes, and the application of such models to oligopolistic competition. The second advance is the proof that logconcavity of the consumer taste density guarantees the existence of a price equilibrium. We analyze an oligopoly model with price competition and free entry, taking explicit account of the integer constraint. Under the Chamberlinian symmetry assumption (that tastes are i.i.d.), we first show that logconcavity of the taste density implies there is excessive market provision of variety when each consumer buys one unit of the product from one of the firms. We then show that this result extends to price-sensitive individual demands by proving that the equilibrium number of firms is at least as great as that which would be provided at the second-best social optimum subject to a zero-profit constraint for firms. Our results call into question previous findings for representative consumer models that left open the possibility of insufficient product diversity.

When selecting amongst a set of investment projects, the decision-maker cannot act as if her decision is made in isolation: Each investment has an impact upon subsequent cash flows and affects with future investments will be feasible and desirable. Our model provides a dynamic context in which her investment decisions can be analyzed. Our model generalizes the multi-period investment models of Gale (1965), Lippman (1970), and Cantor and Lippman (1983), by allowing an arbitrary finite number of projects. We emphasize that neither interest rates, nor net present values, nor internal rates of return are part of our model. However, all three notions arise as a consequence of our basic assumptions. In this paper we introduce a new technique for evaluating bundles of investments, the upper envelope.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at

We consider the allocation of goods in exchange economies with a finite number of agents who may have private information about their preferences. In such a setting, standard allocation rules such as Walrasian equilibria or rational expectations equilibria are not compatible with individual incentives. We characterize the set of allocation rules which are incentive compatible, or in other words, the set of strategy-proof social choice functions. Social choice functions which are strategy-proof are those which can be obtained from trading according to a finite number of pre-specified proportions. The number of proportions which can be accommodated is proportional to the number of agents. Such rules are necessarily inefficient, even in the limit as the economy grows

This note is concerned with nonparanetric and semiparametric inference in regression models where regressors are not continuous. In economic practice, few explanatory variables are continuous. Many of the are dummics, qualitative variables, or counts; and others though continuous in nature, are recorded at intervals and can be treated as discrete.

Two-person repeated games with no discounting are considered where there is uncertainty about the type of the players. If there is a possibility that a player is an automaton committed to a particular pure or mixed stage -game action, then this provides a lower bound on the Nash equilibrium payoffs to a normal type of this player. The lower bound is the best available and is robust to the existence of other types. The results are extended to the case of two-sided uncertainty. This work extends Schmidt (1993) who analysed the restricted class of conflicting interest games.