Allocation with multi criteria decision making techniques

Abstract

Determining the weights of assets in a portfolio is one of the fundamental problems of finance. MCDM techniques are employed for asset allocation purposes. However, criterion weights have to be determined before the steps of the techniques are implemented. In this study, the weights of the criteria are determined by four different MCDM weight techniques (CILOS, CRITIC, MEREC, and SECA), and the effect of these weights on the performance of portfolios created with 17 MCDM techniques (ARAS, CoCoSo, CODAS, COPRAS, EDAS, GRA, MABAC, MAIRCA, MARCOS, MOORA, MOOSRA, OCRA, SAW, TODIM, TOPSIS, VIKOR, WASPAS) is examined. 297 criteria (including sectional and cumulative) were calculated using the mean, standard deviation, and correlation, based on historical returns. The returns of the S&P 500 stocks between January 2020 and December 2021 are used as the dataset. Returns for the first 250 trading days are used to determine the weights of the criteria and the stocks in the portfolio. Returns from the following 250 trading days are used for performance evaluation purposes. The experiment was repeated for two more periods. It is found that cumulative criteria have significantly higher weights than sectional criteria. Differences in returns by industry were also examined. The results show that when the MCDM techniques are used to set criterion weights, a higher return is possible.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 24 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.