Publicação
Risk assessment and management of a portfolio value-at-risk
| Resumo: | This thesis examines the measurement and management of market risk through the application of Value-at-Risk (VaR) methodologies to a diversified portfolio composed of fixed-income securities and equities from the United States, European, and Asian markets. The research pursued two main objectives, namely, to determine which estimation method provides the most reliable VaR and to assess whether a management framework based on this measure can enhance portfolio stability and long-term performance. Several methodologies were considered, including the Normal model, Historical Simulation, Quantile Regression (QR), and the Skewed Generalized Student-t (SGSt) model. The analysis showed that models capable of capturing asymmetrical and heavy tails provide more robust estimates than simpler approaches. Backtesting procedures confirmed the reliability of the chosen model across different market conditions. When applied to portfolio management during 2023 to 2024, the chosen model proved effective in identifying the main sources of risk and guiding the implementation of a dynamic hedging strategy that reduced volatility and improved risk adjusted returns. The contribution of this research lies in its systematic comparison of parametric and non-parametric VaR models within a multi asset portfolio and in demonstrating the practical value of integrating VaR into active portfolio management. The findings demonstrate that diversification on its own is not sufficient in periods of market turbulence, when global markets tend to move together. They underline the importance of combining accurate modelling with timely management actions in order to protect capital and to sustain consistent financial performance over time. |
|---|---|
| Autores principais: | Rosário, João Rodrigues do |
| Assunto: | Value-at-risk Market risk Backtesting Hedging strategy Risk-adjusted returns Portfolio management Risco de mercado Estratégia de cobertura Retornos ajustados ao risco Gestão de portefólio |
| Ano: | 2025 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | ISCTE |
| Idioma: | inglês |
| Origem: | Repositório ISCTE |
| Resumo: | This thesis examines the measurement and management of market risk through the application of Value-at-Risk (VaR) methodologies to a diversified portfolio composed of fixed-income securities and equities from the United States, European, and Asian markets. The research pursued two main objectives, namely, to determine which estimation method provides the most reliable VaR and to assess whether a management framework based on this measure can enhance portfolio stability and long-term performance. Several methodologies were considered, including the Normal model, Historical Simulation, Quantile Regression (QR), and the Skewed Generalized Student-t (SGSt) model. The analysis showed that models capable of capturing asymmetrical and heavy tails provide more robust estimates than simpler approaches. Backtesting procedures confirmed the reliability of the chosen model across different market conditions. When applied to portfolio management during 2023 to 2024, the chosen model proved effective in identifying the main sources of risk and guiding the implementation of a dynamic hedging strategy that reduced volatility and improved risk adjusted returns. The contribution of this research lies in its systematic comparison of parametric and non-parametric VaR models within a multi asset portfolio and in demonstrating the practical value of integrating VaR into active portfolio management. The findings demonstrate that diversification on its own is not sufficient in periods of market turbulence, when global markets tend to move together. They underline the importance of combining accurate modelling with timely management actions in order to protect capital and to sustain consistent financial performance over time. |
|---|