Stock Portfolio Selection Hybridizing Fuzzy Base-Criterion Method and Evidence Theory in Triangular Fuzzy Environment

• Published in: Operations Research Forum Vol. 3; no. 4; p. 53
• Main Authors: Bisht, Kiran, Kumar, Arun
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.12.2022; Springer Nature B.V
• Subjects: Allocations; Applications of Mathematics; Business and Management; Decision making; Fuzzy logic; Fuzzy sets; Investments; Investors; Math Applications in Computer Science; Mathematical and Computational Engineering; Methods; Multiple criterion; Neural networks; Operations Research/Decision Theory; Optimization; Original Research; Portfolio management; Recurrent neural networks; Researchers; Securities markets; Set theory; Standard deviation; Stock exchanges; Stock prices; Subject specialists; India; Stock selection; Triangular fuzzy numbers; Fuzzy Base-Criterion method; Dempster-Shafer evidence theory; Long-short-term memory (LSTM)
• ISSN: 2662-2556; 2662-2556
• DOI: 10.1007/s43069-022-00167-3

In this paper, an integrated multi-criteria decision-making framework is developed for portfolio construction by unifying the assessments of a novice investor and stock market expert. This paper aims to simplify the stock selection process for inexperienced investors with limited understanding. The critical factors of portfolio construction have been selected based on novice investor’s knowledge and domain expert’s extensive analysis. The weights of the stock selection criterion are determined using the fuzzy Base-Criterion method. The Dempster-Shafer evidence theory is used to collect and combine evidence to categorize securities as “good for investment,” “average for investment,” or “bad for investment.” The securities are ranked based on the consensus decision of both decision-makers. A portfolio of ten top-ranked securities is formed. The Sharpe ratio is optimized using a deep recurrent neural network with long-short-term memory (LSTM) to obtain optimal ratio allocations of stocks in the portfolio. The proposed model is applied for investment in the National Stock Exchange of India. The effectiveness of the proposed model is verified by sensitivity and comparative analysis.