[This paper estimates all possible multidimensional interaction effects in a logit model of homeownership, assesses the relative importance of these interactions, and interprets the results in light of existing theories of housing consumption.]

A wide variety of economic models include as explanatory variables either expectational variables or variables representing the result of some decision-making process. The first category includes both expectations about the future values of variables, e.g., next period's sales, the level of unemployment two quarters ahead, etc. and other subjective variables such as permanent income or the "normal" level of prices and interest rates. Examples of the second type are "desired" capital stock, planned production, or inventory accumulation, and so on.

This paper explores a new approach to the Nash bargaining problem in which the axiom of symmetry is dropped and it is assumed that the final allocation depends on both the status quo and the threat point. The resulting final allocation, unlike that formalized by Nash, cannot be represented by a simple analytic expression; rather, it leads to a whole class of solutions. Properties of the final allocation are analyzed. It is shown that for every initial allocation there exists a Nash fiber, corresponding to the Nash allocation, that it is possible to determine the sign of the derivatives of the final allocation with respect to changes in the threat point, and that a Slutsky-like equation relates these derivatives to the derivatives with respect to the initial allocation. It is also shown that, under certaiia conditions, as play is repeated the final allocation asymptotically converges to the Nash allocation.

[The sets of local demand functions which would generate either linear or nonlinear schedules in a heterogenous as well as homogeneous space economy are determined by iterative applications of the separation of variables technique for solving differential equations. Nonlinear delivered price schedules which reflect profit maximization objectives are thereby identifiable. In turn, these schedules can be distinguished from those which involve strictly predatory price behavior.]