This work discusses the Sensitivity Analysis (SA) of portfolio volatility ( σ_{p}) and its role in the interpretation of trading/reallocation strategies. Starting from recent findings in the SA field, we show that results obtained utilizing partial derivatives (PD) or Elasticity (E) cannot be applied to the analysis of the generic trading strategy. We show that such limitations can be overcome by making use of the Differential Importance Measure (D). We also show that, thanks to D additivity property, the importance of groups of weights (Sectors) can be derived straightforwardly. This result enables us to analyze trading/reallocation strategies in terms of portfolio composition providing quantitative information on diversification across sectors. After obtaining the expression of D for individual weights with respect to σ_{p} estimated via GARCH models, numerical results are discussed for a portfolio of 30 stocks composing the Dow Jones Index. We show that quantitative comparison of the agreement among alternative reallocation strategies can be found by applying Savage Score Correlation Coefficients to results produced by D. Numerical information on portfolio diversification across sectors is finally discussed.
Sensitivity analysis of Portfolio Volatility: Importance of Weights, Sectors and the Impact of Trading Strategies
BORGONOVO, EMANUELE;PERCOCO, MARCO
2007
Abstract
This work discusses the Sensitivity Analysis (SA) of portfolio volatility ( σ_{p}) and its role in the interpretation of trading/reallocation strategies. Starting from recent findings in the SA field, we show that results obtained utilizing partial derivatives (PD) or Elasticity (E) cannot be applied to the analysis of the generic trading strategy. We show that such limitations can be overcome by making use of the Differential Importance Measure (D). We also show that, thanks to D additivity property, the importance of groups of weights (Sectors) can be derived straightforwardly. This result enables us to analyze trading/reallocation strategies in terms of portfolio composition providing quantitative information on diversification across sectors. After obtaining the expression of D for individual weights with respect to σ_{p} estimated via GARCH models, numerical results are discussed for a portfolio of 30 stocks composing the Dow Jones Index. We show that quantitative comparison of the agreement among alternative reallocation strategies can be found by applying Savage Score Correlation Coefficients to results produced by D. Numerical information on portfolio diversification across sectors is finally discussed.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
