Article Abstract:
Sensitivity analysis determines effectiveness of multi-criteria decision making methods (MCDM) in solving decision problems. MCDM methods, such as the weighted sum model (WSM), weighted product model (WPM), and the analytic hierarchy process (AHP) are used to derive critical results of sensitivity analysis on a decision problem given a number of alternatives and decision criteria. A study of this sensitivity analysis approach shows a common pattern of performance of all three MCDM methods in a computational experiments. The choice of an MCDM method or combination of methods is found to have no significant impact on sensitivity results.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
Research to validate an assumption of multivariate normality for independent variables from a bond rating decision model is presented. Five variables are examined for both multivariate and univariate normality. The nonnormal variables receive normalizing transformations. Three methods are used to investigate the variables. Computer graph plots are provided. Research results validate the original assumption. Investigative methods can be used in other decision models.
User Contributions:
Comment about this article or add new information about this topic:
Article Abstract:
A new sensitivity analysis strategy incorporating dependence relations among the input variables using rank correlations is presented. This technique generates affine combinations of input variables that are approximately pairwise uncorrelated.
User Contributions:
Comment about this article or add new information about this topic: