Model-building professionals are often facing a very difficult choice of selecting relevant variable/s from a set of several similar variables. All those variables are supposedly representing the same factor but are measured differently. They are based on different methodologies, baselines, conversion/comparability methods, etc., thus leading to substantial differences in numerical values for essentially the same things (from the perspective of the modeler).
In this study, we introduce a method for modeling, capable of utilizing ranges (intervals) of values and thus enabling to utilize inclusive approach, which includes all the relevant variables that represent the same factor. This approach has numerous advantages in terms of efficient data utilization, reliability of conclusions and simplification of process to summarize results (due to reduction in the number of necessary regression runs). We also introduce an interval reduction algorithm, designed to reduce excessive size of intervals, thus bringing them closer to their central tendency cluster. The algorithm also allows to identify exceptional cases where such reduction is not possible.
The modeling method used in this study is Soft Regression (based on Fuzzy Logic).
- SELECTION OF COCOA POST-HARVEST TECHNOLOGY USING FUZZY LOGIC
- AN EXTENSION ON FUZZY INTEGER LINEAR PROGRAMMING WITH EQUALITY CONSTRAINS
- Customer loyalty in services: an approximation by means of fuzzy logic
- Computing competencies value through a fuzzy real option model
- Uncertainty Theory Applied to Optimal Selection of Personnel in an Enterprise