Normalization Technique Selection for MCDM Methods: A Flexible and Conjunctural Solution that can Adapt to Changes in Financial Data Types

Authors : Mahmut Baydaş, Tevfik Eren, Mustafa Iyibildiren

Pages : 148-164

View : 72 | Download : 119

Publication Date : 2023-12-17

Article Type : Research Paper

Abstract :It makes sense to use the MCDM methodology to select and rank alternatives for a multi-criteria problem. As it is known, since it is not possible to use criteria consisting of different units in a common calculation, it is necessary to convert them into a unitless dimension. Many alternative normalization techniques have been proposed in the past for this conversion process. On the other hand, normalization techniques that appear to be accurate fair transformers have the potential to affect the final ranking of any MCDM method, and this is a significant problem. As a matter of fact, these alternative techniques can change the best alternative and overall ranking for an MCDM. Therefore, an unconsciously chosen normalization technique may reduce the quality of the findings. However, it cannot be said that there is a consensus on the choice of normalization methods. Previous studies have unanimously stated that normalization methods can affect MCDM results. In this study, from an innovative perspective, the effect of normalization methods on the results is evaluated with a third party, an external constant factor. In other words, we focus on how the normalization technique affects the relationship of MCDM with an external factor. Thus, we want to achieve a fair assessment by choosing a reference point. According to the findings of the approach tested in different financial data sets, the most successful technique may change periodically. Therefore, the selection of the best normalization technique depending on the data structure should be evaluated from a dynamic rather than static perspective.
Keywords : Normalizasyon, ÇKKV, Finansal Veri Analitiği