Portfolio optimization in fuzzy asset management with coherent risk measures derived from risk averse utility

A portfolio optimization problem with fuzzy random variables is discussed using coherent risk measures, which are characterized by weighted average value-at-risks with risk spectra. By perception-based approach, coherent risk measures and weighted average value-at-risks are extended for fuzzy random...

Full description

Saved in:
Bibliographic Details
Published inNeural computing & applications Vol. 32; no. 15; pp. 10847 - 10857
Main Author Yoshida, Yuji
Format Journal Article
LanguageEnglish
Published London Springer London 01.08.2020
Springer Nature B.V
Subjects
Artificial Intelligence
Asset management
Coherence
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Science
Data Mining and Knowledge Discovery
Evaluation
Image Processing and Computer Vision
Mathematical programming
Optimization
Probability and Statistics in Computer Science
Random variables
Randomness
Risk perception
S.I. : India Intl. Congress on Computational Intelligence 2017
Utility functions
Portfolio optimization
Perception-based extension
Fuzzy random variable
Coherent risk measure
Weighted average value-at-risk
Online AccessGet full text

Cover

More Information
Summary:A portfolio optimization problem with fuzzy random variables is discussed using coherent risk measures, which are characterized by weighted average value-at-risks with risk spectra. By perception-based approach, coherent risk measures and weighted average value-at-risks are extended for fuzzy random variables. Coherent risk measures derived from risk averse utility functions are introduced to discuss the portfolio optimization with randomness and fuzziness. The randomness is estimated by probability, and the fuzziness is evaluated by lambda-mean functions and evaluation weights. By mathematical programming approaches, a solution is derived for the risk-minimizing portfolio optimization problem. Numerical examples are given to compare coherent risk measures. It is made clear that coherent risk measures derived from risk averse utility functions have excellent properties as risk criteria for these optimization problems. Not only pessimistic and necessity case but also optimistic and possibility case are calculated numerically to deal with uncertain information.
ISSN:0941-0643
1433-3058
DOI:10.1007/s00521-018-3683-y