Showing papers by "Abraham Charnes published in 1976"

The equivalence of generalized least squares and maximum likelihood estimates in the exponential family.

Abstract: The method of iterative weighted least squares can be used to estimate the parameters in a nonlinear regression model. If the dependent variables are observations from a member of the regular exponential family, then under mild conditions it is shown that the IWLS estimates are identical to those obtained using the maximum likelihood principle. An application is provided to illustrate the results.

Coalitional and Chance-Constrained Solutions to N-Person Games. I. The Prior Satisficing Nucleolus.

Abstract: In this report we define and study a “prior satisficing nucleolus”—a solution concept for n-person cooperative games in stochastic characteristic function formWe provide two different proofs for the existence of this “prior” nucleolus, and prove its uniqueness for a wide class of distribution functions

A theorem on homogeneous functions and extended Cobb-Douglas forms.

Abstract: A form for homogeneous functions is presented which shows them to be a very simple extension of the wellknown Cobb-Douglas functions with similar properties in production (and distribution) economics. This form thus suggests new possibilities for interpreting a wide variety of empirical and theoretical results in economics; it also provides contact with developments in other fields, such as information theory and geometric programming.

A Multi-Objective Model for Planning Equal Employment Opportunities.

Abstract: Previous manpower planning models — e.g., in the OCMM series — have utilized multi-period Markoff processes embedded in goal programming (multiple objective) models. These are here extended to Equal Employment Opportunity plans directed to changing the mix of employees over time. At each point in the planning interval, the organization is taken as given, e.g., in terms of the probabilities for promotion, transfer, etc., When formulating manpower programs. Over time, however, these organization processes are submitted to planned changes which alter the probabilities of occurrence for these events. The Merit Promotion System is preserved and other controls are also imposed explicitly for the exercise of managerial discretion. The focus here is on an ordinary (absolute value) formulation of objectives and a numerical illustration is supplied with differing weights for each of the indicated classes of objectives. Other types of objectives are briefly discussed, along with different approaches to problems of validation and, subsequently, implementation in a U.S. Navy context.

A futurological justification for historical cost and multi-dimensional accounting

Abstract: Futurology, interpreted as a forecast or, even better, as projections intended to influence the future, is applied to justify retention of historical costs in the context of proposed multidimensional accounting extensions for business, government and national goals accounting. Budgetary disclosures supported by opportunity costs and tradeoffs are also examined via multiple-objective models of goal programming varieties. Independent audit and attest is suggested not only for the mathematical model but also for managerial explanations of subsequent discrepancies from plans in each pertinent dimension.

An algorithm for solving general fractional interval programming problems

Abstract: A Linear Fractional Interval Programming problem (FIP) is the problem of extremizing a linear fractional function subject to two-sided linear inequality constraints. In this paper we develop an algorithm for solving (FIP) problems. We first apply the Charnes and Cooper transformation on (FIP) and then, by exploiting the special structure of the pair of (LP) problems derived, the algorithm produces an optimal solution to (FIP) in a finite number of iterations.

A note on explicit solution in linear fractional programming

Abstract: In this note we analyze the fractional interval programming problem (FIP) and find, explicitly, all its optimal solutions. Though our results are essentially the same as those in Charnes and Cooper [4], the proofs and analysis we provide here are considerably simpler.