Performance of Different Risk Indicators in a Multi-Period Polynomial Portfolio Selection Problem Based on the...

Area of Science:

• Quantitative Finance
• Investment Management
• Operations Research

Background:

• Selecting appropriate portfolio selection models is crucial for non-professional investors.
• Traditional models may not adequately address multi-period investment horizons and diverse risk indicators.
• Transaction costs are a significant factor in multi-period investment strategies.

Purpose of the Study:

• To formulate a multi-period portfolio selection model incorporating transaction costs.
• To integrate multiple risk indicators (variance, semi-variance, entropy, semi-entropy) using a credibility measure.
• To provide a framework for investors to choose models aligned with their risk preferences.

Main Methods:

• Formulation of a multi-period polynomial portfolio selection model based on credibility measures.
• Application of the polynomial goal programming (PGP) approach to combine risk indicators.
• Design of an adjusted genetic algorithm with a penalty function for optimization.

Main Results:

• The PGP method effectively integrates various risk indicators for portfolio selection.
• The developed model successfully handles multi-period investment scenarios with transaction costs.
• The adjusted genetic algorithm efficiently finds optimal solutions for the complex model.

Conclusions:

• The PGP approach is suitable for investors in selecting appropriate portfolio models.
• The study provides a flexible framework for investors with varying risk appetites.
• The model aids in making informed investment decisions in multi-period contexts.