Knowledge that Transforms

This paper proposes the residual-based stochastic predictor as an alternative procedure for obtaining forecasts with a static nonlinear econometric model. This procedure modifies the usual Monte Carlo approach to stochastic simulations of the model in that calculated residuals over the sample period are used as proxies for disturbances instead of random draws from some assumed parametric distribution. In compar-ison with the Monte Carlo predictor, the residual-based should be less sensitive to distributional assumptions concerning disturbances in the system. It is also less demanding computationally. The large-sample asymptotic moments of the residual-based predictor are derived in this paper and compared with those of the Monte Carlo predictor. Both procedures are asymptotically unbiased. In terms of asymptotic mean squared prediction error (AMSPE), the Monte Carlo is efficient relative to the residual-based when the number of replications in the Monte Carlo simulations is large relative to sample size. This order of relative efficiency is reversed, however, when replication and sample sizes are similar. In any event, the amount by which the AMSPE of either predictor exceeds the lower bound for AMSPE is small as a percentage of the lower bound AMSPE when sample and replication sizes are at least of moderate magnitude. The paper also discusses the extension of the residual-based anld Monte Carlo procedures to the estimation of higher order moments and cumulative distribution functions of endogenous variables in the system.

[This paper develops a unified framework for formulating econometric models of discrete/continuous consumer choices in which the discrete and continuous choices both flow from the same underlying (random) utility maximization decision. As a special case a number of models suitable for empirical application are developed where the discrete choice is among different brands of a commodity. Since these brands are essentially substitutes, the consumer prefers to buy only one brand at any time; discrete choice is which brand to select and the continuous choice is how many units to buy.]

The paper uses a generalized Nash bargain to analyze input levels, profits, and wages in the absence of binding contracts, and compares these with the convenitional binding contracts model. It is shown that if the union has any power, investment is lower in the absence of binding contracts. The associated input levels and shareholders' profits are identical to those that emerge if contracts are binding and the firm acts as if it faces a cost of capital which is a linear combination of the purchase price of capital and the resale value. This implicit cost is greater than the purchase price, is an increasing function of union power, and is independent of the profit function and the alternative wage. Increases in union power reduce shareholders' profits but may increase wages at some points and decrease wages at others. In the absence of binding contracts, shareholders' profits are lower but there is a critical level of union power (depending on the profit function) such that the union is worse off if its power is higher than this level and better off if it is lower.

This paper focuses on developing and adapting statistical models of counts (nonnegative integers) in the context of panel data and using them to analyze the relationship between patents and R & D expenditures. Since a variety of other economic data come in the form of repeated counts of some individual actions or events, the methodology should have wide applications. The statistical models we develop are applications and generalizations of the Poisson distribution. Two important issues are (i) Given the panel nature of our data, how can we allow for separate persistent individual (fixed or random) effects? (ii) How does one introduce the equivalent of disturbances-in-the-equation into the analysis of Poisson and other discrete probability functions? The first problem is solved by conditioning on the total sum of outcomes over the observed years, while the second problem is solved by introducing an additional source of randomness, allowing the Poisson parameter to be itself randomly distributed, and compounding the two distributions. Lastly, we develop a test statistic for the presence of serial correlation when fixed effects estimators are used in nonlinear conditional models.

[An econometric problem in estimating models of occupational choice is that the agents' forecasts of future wages and occupational tenure are unobservable. This paper solves the problem by assuming that agents have rational expectations and by considering the effects of arbitrage both within and between cohorts. The solution consists of two time series regressions of the demand and supply functions of entrants into an occupation. From these regressions one obtains estimates of the rate of return to education, the direct cost of education, and other parameters that influence the market. The model was estimated with data from the market for lawyers.]

[Discrete choice models are now used in a variety of situations in applied econometrics. By far the model specification which is used most often is the multinomial logit model. Yet it is widely known that a potentially important drawback of the multinomial logit model is the independence from irrelevant alternatives property. While most analysts recognize the implications of the independence of irrelevant alternatives property, it has remained basically a maintained assumption in applications. In the paper we provide two sets of computationally convenient specification tests for the multinomial logit model. The first test is an application of the Hausman [10] specification test procedure. The basic idea for the test here is to test the reverse implication of the independence from irrelevant alternatives property. The test statistic is easy to compute since it only requires computation of a quadratic form which involves the difference of the parameter estimates and the differences of the estimated covariance matrices. The second set of specification tests that we propose is based on more classical test procedures. We consider a generalization of the multinomial logit model which is called the nested logit model. Since the multinomial logit model is a special case of the more general model when a given parameter equals one, classical test procedures such as the Wald, likelihood ratio, and Lagrange multiplier tests can be used. The two sets of specification test procedures care then compared for an example where exact and approximate comparisons are possible.]

This paper proposes a procedure for implementing efficient egalitarian equivalent allocations in an exchange economy, using the perfect equilibrium concept. This procedure is an extension of the divide and choose method in two ways: it is defined for more than two agents and the divider's advantage is removed by auctioning the role of divider among the agents (as in Crawford [1]).Thus, in contrast with other equilibrium concepts (Nash, dominant), the perfect one solves the efficiency-justice dilemma.

[This paper considers approximations to the distribution of the least squares estimator of α in the model @y"t = @a@y"t"-"1 + @u"t where the @u"t are independently distributed N(0, @s extasciicircum2) and @y"0 is fixed. An Edgeworth approximation for this case is calculated, and compared with results for the stationary case. For extbar@a extbar extgreater 1 and fixed @y"0, the asymptotic distribution is found in closed form; when extbar@a extbar extgreater 1 and @y"0 = 0, an Edgeworth-type approximation is again calculated; this is compared with exact results.]

A NUMBER OF AUTHORS argue that a Bayesian posterior odds criterion is appropriate for model selection.2 This paper considers how to derive this criterion when there is minimal prior information. We propose minimizing measures of prior information relative to the models in question rather than relative to the parameters of the particular models. In so doing, we obtain an expression for the odds that is invariant to the parameterization of the particular models and overcomes certain well known finite sample limiting problems. We illustrate this procedure using two popular measures of information derived from the well known Shannon [26] measure. By minimizing these measures with the sample size held fixed, we obtain the same model selection criterion that Schwarz [25] derived asymptotically for large sample sizes. This expression has a number of desirable properties and is computationally no more