Regime Switching Models and the Mental Accounting Framework

Inhaltsverzeichnis (Table of Contents)

• Introduction
• Literature Review
  • The Mental Accounting Framework
    • Markowitz's MVPT vs. MA framework.
    • Mean-Variance Optimization of Mental Accounts
  • Dynamic Investment Management
    • Components of a Stochastic DP model
    • Abstraction of the stochastic programming approach
    • Scenario Generation
  • Regime Switching and Markov Chain Models
    • The Three Basic Problems for HMMs
    • Components of a Hidden Markov Model
    • Model Selection.
  • Gaussian Mixture Models
    • Definition of Gaussian Mixtures
    • Gaussian Mixture Models in Stochastic Programming
    • Moments of a Gaussian Mixture Distribution
• Mental Accounting with Regime Switching
  • Stochastic Programming Problem Formulation
    • Objective Function
    • Constraints
    • Definition of Stages.
  • Scenario Generation
  • Decision Policy
  • Backtesting
  • Hypotheses
• Market Data and Methodology
  • Market Data
  • Methodology
  • Missing Data.
• Results
  • HMM Calibration
    • Model Selection.
    • Analysis of the selected model
  • Scenario Generation
  • Asset Allocation.
  • Backtesting.
• Summary and Conclusion

Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)

This Master thesis investigates the application of regime switching models to the mental accounting framework in portfolio optimization. The study aims to incorporate the dynamic nature of market conditions and investor behavior into asset allocation strategies.

• Mental Accounting Framework
• Regime Switching Models
• Dynamic Investment Management
• Stochastic Programming
• Backtesting and Performance Evaluation

Zusammenfassung der Kapitel (Chapter Summaries)

The thesis begins with a review of relevant literature on mental accounting and dynamic investment management, exploring the relationship between investor behavior and portfolio construction. It then introduces regime switching models, particularly Hidden Markov Models (HMMs), and Gaussian Mixture Models (GMMs), as tools for capturing market dynamics. The main chapters delve into the application of these models to the mental accounting framework, outlining the problem formulation, scenario generation, decision policy, and backtesting procedures. The final chapters present the results of the analysis, including model calibration, scenario generation, asset allocation strategies, and backtesting performance evaluations.

Schlüsselwörter (Keywords)

The thesis focuses on the intersection of behavioral finance, quantitative finance, and dynamic asset allocation. Key keywords include Mental Accounting, Regime Switching Models, Hidden Markov Models, Gaussian Mixture Models, Stochastic Programming, Portfolio Optimization, Backtesting, and Risk Management.