Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
INCOMPLETE MODEL SPECIFICATION AND SENSITIVITY ANALYSIS USING CAUSAL MODELING (C-V-P, INFORMATION, DECISION MAKING)
Abstract
The focus of this research is the investigation of the firm's short-run decision making and the information problems associated with them. Its basic premise is that the firm's decision making can be improved by integrating causal theory and path analysis into human information processing and judgement. Three objectives were established for this research: (1) develop a theoretical method of incomplete model specification for a complex stochastic process. (2) provide information concerning the sensitivity of the model to measurement error. (3) demonstrate the theory's robustness. The theoretical methods of incomplete model specification and sensitivity analysis were developed by demonstrating that short-run decisions of the firm were best described by the incomplete information economic model. It was then shown that the elements of the incomplete model were selected using heuristic methods. Heuristics were proven to be poor selectors of optimal decision models by a review of the current human information processing research. Causal modeling was then proposed as an aid for the selection of the decision model's elements. The evolutionary development of the information system was summarized by an iterative decision model. The model consisted of three levels of decision making: a low level implementation decision, an intermediate level decision concerning the maximization of the objective function, and a high level specification of the decision model's elements. The specialization of judgement tasks resulted in improved information utilization and selection. The feasibility of the iterative decision model was demonstrated by applying it to an empirical example. The intermediate level decision was obtained by a surface search over the objective function. The model specification decisions were made using causal theory and path analysis. These techniques were used to determine the causal strength and statistical significance of each variable in the empirical example. The sensitivity of the decision model to measurement error was accomplished by decomposing the correlation between each predictor variable and the objective function. This decomposition was used to calculate the potential damage due to adverse selection or moral hazard.
Subject Area
Accounting
Recommended Citation
BROMLEY, ROBERT GENE, "INCOMPLETE MODEL SPECIFICATION AND SENSITIVITY ANALYSIS USING CAUSAL MODELING (C-V-P, INFORMATION, DECISION MAKING)" (1984). ETD collection for University of Nebraska-Lincoln. AAI8427898.
https://digitalcommons.unl.edu/dissertations/AAI8427898