[This paper describes the U.S. offshore oil and gas lease sales conducted by the Department of the Interior since 1954. Several decisions are discussed, including bidding for leases, the government's decision whether to accept the highest bid, the incidence and timing of exploratory drilling, and the formation of bidding consortia. Equilibrium models of these decisions that emphasize informational and strategic issues and that account for institutional features of the leasing program are analyzed, and their predictions compared to outcomes in the data.]

In this paper, we develop a classical approach to model selection. Using the Kullback-Leibler Information Criterion to measure the closeness of a model to the truth, we propose simple likelihood-ratio based statistics for testing the null hypothesis that the competing models are equally close to the true data generating process against the alternative hypothesis that one model is closer. The tests are directional and are derived successively for the cases where the competing models are non-nested, overlapping, or nested and whether both, one, or neither is misspecified. As a prerequisite, we fully characterize the asymptotic distribution of the likelihood ratio statistic under the most general conditions. We show that it is a weighted sum of chi-square distribution or a normal distribution depending on whether the distributions in the competing models closest to the truth are observationally identical. We also propose a test of this latter condition.

Time series variables that stochastically trend together form a cointegrated system. OLS and NLS estimators of the parameters of a cointegrating vector are shown to converge in probability to the true parameter value at the rate T11d for any positive d. These estim mators can be written asymptotically in terms of relatively simple nonnormal random matrices which do not depend on the parameters of th e system. These asymptotic representations form the basis for simple and fast Monte Carlo calculations of the limiting distributions of th ese estimators. Asymptotic distributions thus computed are tabulated for several cointegrated processes.

[The formulation of optimal policy in linear rational expectations models is studied using methods analogous to the classical design techniques utilized in linear systems engineering. Specifically, the policy-maker's present-value-like objective function is converted, using the convolution transform, to an equivalent frequency domain, "spectral utility" function. Then the residue calculus and Wiener-Hopf methods are used to maximize spectral utility through the choice of a complex function which represents a sequence of distributed lag coefficients to be applied to current and past values of instrument variables. The solution to this problem is a closed form expression for the decision rule of the dominant player in a particular type of linear-quadratic dynamic game.]

[Under the assumption that demand behavior depends on intertemporal preferences as well as (point) expectations concerning future prices, it is demonstrated that under plausible conditions rationality imposes no observable restrictions on the demand function and expectations and preferences are observationally indistinguishable.]

[The Arrow-Debreu model is extended to include a sequential market model with financial markets. This is done by dropping the contingent contracts from the Arrow-Debreu model, leaving only a sequence of spot markets for commodities. The resulting market structure is inefficient. Efficiency is restored with a sequence of stock markets and option markets. In addition, consumers are shown to be unanimous in wanting each firm to maximize the price of its common stock.]

[Techniques are described whereby the distribution of completed unemployment spell lengths may be inferred from the distribution of in-process unemployment spell lengths recorded each month in the Current Population Survey. An extension is proposed whereby the complete population joint distribution of labor market transition probabilities can be estimated using only Current Population Survey information.]

[This paper presents a Bayesian analysis of a single equation from a simultaneous equations system. The analysis is carried out under "limited information" because no prior information (other than a list of endogenous and exogenous variables) is introduced on the parameters of the remaining equations in the in the system. These parameters are integrated out analytically. The equation of interest may or may not be identified by means of exact a priori information; probabilistic prior information is equally acceptable. The prior density is either of the non-informative or the natural conjugate type. The kernel of the posterior density for the regression coefficients is a ratio of t kernels. The existence of posterior moments is ascertained. This approach is applied for illustrative purposes to Tintner's model of the meat market.]

The impatience implications of continuous time utility indicators are interesting to the extent that they differ from the discrete time results. The class of tFaditional integral utility indicators are considered and impatience implications are shown to depend on the dif- ferent convergence implications of the continuous time case. The stronger separability assumptions of continuous time utility indicators allow a weakening of compactness assumptions often required to demonstrate impatience. presence of impatience. Specific separability assumptions were invoked by Koopmans (8) and Koopmans, Diamond, and Williamson (9) in order to demonstrate the presence of impatience in problems involving choice over an infinite program horizon. From a paper by Diamond (4) one,can infer much of the relationship between separability assumptions and impatience implications. Diamond employed several intertemporal non-complementary assumptions to demonstrate eventual impatience for a case in which the consumption space was not compact in the topology of the norm. The use of non-complementary axioms seems justifijable as their economic implications are straightforward while those of compactness assumptions are not immediately obvious.2 Moreover the natural extension of Diamond's first axiom to all time periods yields a condition equivalent to the independence assumption employed by Debreu (3) in representing preferences by an additive function. Consequently, this paper analyzes separable utility indicators directly for impatience implications; the analysis considers the continuous time case as it subsumes the discrete time analog. However the discrete time case will be discussed in order to facilitate analogy construction.