EMANUELE BORGONOVO AND MARCO PERCOCO 67- The discussion of the relationship between DIM and the Fussel–Vesely importance
can be found in Borgonovo and Apostolakis (2001a), the discussion on the relation-
ship between DIM and Elasticity can be found in Borgonovo and Peccati (2004) and
Borgonovo and Peccati (2005). - Geometrically this would correspond to considering the direction of the change, not
only its projection on the cartesian axes. - Some technical notes on computation. The optimal strategy in Manganelli (2004)
implies the changeda∗=a∗−a^0 , wherea^0 is the initial point. More specifically, we
consider as a starting pointa^0 the result of the estimation of the weights for a given
portfolio as defined by the first-order conditions of an exponentially weighted mov-
ing average (EWMA) (Manganelli, 2004). The choice of that particular stochastic
process as a generator of the weights has been made as Manganelli (2004) demon-
strates that the volatility of the 30 stocks estimated by an EWMA is only 7.34% lower
than the one estimated by a GARCH(1,1) model ina∗, wherea∗is the vector of
optimal weights minimizing the variance of the portfolio. This result is particularly
interesting for our purposes as the EWMAoptimal portfolio can be thought as a local
deviation, in terms of volatility, from the GARCH(1,1) computed in the optimum.
Furthermore, Following Wang’s (1998) arguments, we do not impose any constraint
on portfolio weights as to run the analysis under maximum attainable efficiency.
REFERENCES
Black, F. and Scholes, M. (1973) “The Pricing of Options and Corporate Liabilities”,Journal
of Political Economy, 81(3): 637–54.
Bollerslev, T. (1986) “Generalized Autoregressive Conditional Heteroskedasticity”,Jour-
nal of Econometrics, 31: 307–27.
Bollerslev, T. and Engle, R.F. (1993) “Common Persistence in Conditional Variances”,
Econometrica, 61(1): 167–86.
Borgonovo, E. (2001) “Risk-Informed Decision Making: Sensitivity Analysis and Appli-
cations”, Ph.D thesis, Massachusetts Institute of Technology.
Borgonovo, E. and Apostolakis, G.E. (2001) “A New Importance Measure for Risk-
Informed Decision-Making”,Reliability Engineering and System Safety, 72(2): 193–212.
Borgonovo, E. and Apostolakis, G.E. (2001) “On Local Sensitivity Measures”,Proceedings
of Sensitivity Analysis of Model Output, Madrid, Spain.
Borgonovo, E. and Peccati, L. (2004) “Sensitivity Analysis in Investment Project Evalua-
tion”,International Journal of Production Economics, 90: 17–25.
Borgonovo, E. (2001) “Importance Relations and SensitivityAnalysis in Decision Theory”,
Proceedings of the Tenth International Conference on Foundations and Applications of Utility,
Risk and Decision Theory, Torino, 30 May–2 June 2001, pp. 20–3.
Borgonovo, E. and Peccati, L. (2004) “The Importance of Assumptions in Investment
Evaluation”,International Journal of Production Economics, 90: 17–25.
Campbell, J.Y., Lo, A.W. and MacKinlay, A.C. (1997)The Econometrics of Financial Markets
(Princeton, NJ: Princeton University Press).
Campolongo, F. and Saltelli, A. (1997) “Sensitivity Analysis of an Environmental Model:
anApplication of DifferentAnalysis Methods”,Reliability Engineering and System Safety,
57: 49–69.
Cheok, M.C., Parry, G.W. and Sherry, R.R. (1998) “Use of Importance Measures in Risk-
Informed RegulatoryApplications”,Reliability Engineering and System Safety, 60: 213–26.
Drudi, F., Generale, A. and Majnoni, G. (1997) “Sensitivity of VaR Measures to Different
Risk Models”, Banca d’Italia,Temi di discussione, no. 317.