위험척도별 균형위험 포트폴리오 전략 분석

This study conducts an empirical analysis of Risk Parity (RP) portfolio strategies in the Korean equity market, focusing on various risk measures, including Volatility, Value-at-Risk (VaR), Conditional Value-at-Risk (CVaR), and Maximum Drawdown (MDD). The portfolios were constructed based on both total risk and residual (idiosyncratic) risk, using quintile-based and risk parity weighting schemes. The performance of these strategies was compared against value-weighted and equal-weighted market portfolios over different rebalancing cycles (six-month and one-year). The results yield several key insights. First, predefined risk measures exhibited strong predictive power for ex-post realized risk, supporting the validity of risk-based portfolio construction. Notably, the ordering of portfolios based on ex-ante risk estimates was consistent with observed risk outcomes, even across alternative risk metrics. This consistency reinforces the robustness of risk measure-based classification for asset allocation. Second, meaningful cross-predictive power was observed across different risk measures. For instance, portfolios sorted by CVaR also exhibited systematic ordering in terms of MDD and VaR, suggesting the existence of shared informational content among these risk metrics. This finding implies that multidimensional risk management, incorporating multiple measures concurrently, may be more effective than relying on a single risk dimension. Third, in terms of returns, RP portfolios based on VaR and CVaR generated superior average returns and lower volatility compared to traditional benchmarks. Interestingly, while MDD-based portfolios underperformed under total risk settings, they showed statistically significant performance improvements when residual risk was used as the basis. This result indicates that downside-focused risk measures like MDD may be more effective when applied to idiosyncratic volatility rather than market-driven total risk. Fourth, the length of the rebalancing cycle significantly influenced strategy effectiveness. The six-month rebalancing schedules yielded stronger excess returns and statistical significance compared to one-year cycles. This suggests that RP strategies require dynamic reallocation to maintain their risk-targeting efficacy, particularly in fast-changing market environments. These findings collectively demonstrate the practical viability of Risk Parity strategies in the Korean stock market, with important implications for asset managers seeking more stable and risk-aware allocation frameworks. Beyond portfolio construction, the study also highlights the potential for enhanced downside protection and return efficiency through nuanced applications of risk metrics. Nevertheless, the study has certain limitations. The analysis assumes frictionless trading and excludes transaction costs, taxes, and liquidity constraints. Additionally, the study is confined to equities, leaving the applicability of RP strategies to multi-asset portfolios unexplored. Future research could expand the current framework by integrating alternative asset classes, incorporating real-world constraints, or utilizing machine learning techniques to enhance risk forecasting and allocation precision. In conclusion, this study enriches the academic and practical understanding of risk-based investing by providing empirical evidence on the performance of Risk Parity portfolios constructed under diverse risk metrics in an emerging market context. It also emphasizes the need for multidimensional, dynamically managed risk frameworks in modern portfolio management.