Search

We develop a framework for understanding how variations of portfolio properties are apportioned to changes in portfolio composition as trading is performed in the presence of budget constraints. Our approach is based on the concept of constrained derivative. It allows one to obtain the simultaneous sensitivity of portfolio properties w.r.t. assets groups for any choice of pivotal asset. Analytical expressions for the sensitivity of portfolio returns, variance, GARCH volatility and Value at Risk (VaR) are derived. A numerical exemplification is proposed with reference to the 30 assets of the Dow Jones Index.

The representation of multivariate functions is often given by means of functional ANOVA, in particular in Statistics, Simulation and Sensitivity Analysis. In despite of such use, its analytic and geometric properties have not been addressed yet. We derive conditions for continuity, differentiability, monotonicity and ultramodularity properties in functional ANOVA. We study the implications of these findings in multiattribute utility theory. We establish the conditions under which analytic properties of a multiattribute utility function imply the same properties of its one-attribute functions. The converse implications are also investigated.

The lack of analytical test cases is one of the open issues in moment independent SA. In this work, by exploiting results in a previous paper by the authors, we present several analytical test cases. We also run corresponding numerical computations that corroborate the analysis.

Recent literature has recognized the importance of including interactions in reliability importance measures. However, a study of when interactions do matter in complex PSA models and how they relate with uncertainty has not been performed yet. It is the purpose of this work to investigate such relationship. In order to perform our investigation, we make use of the recently introduced total order reliability importance measures. Applications are obtained via a space PSA model.

In this paper we apply a moment-independent sensitivity analysis technique to the Inoperability Input-Output Model to identify the most effective preparedness strategies.

The present issue contains the abstracts of the works presented at the Sixth International Conference on Sensitivity Analysis of Model output, held in Milan, at Bocconi University from July 19th to July 22nd 2010. The conference has been co-organized by the ELEUSI research center of Bocconi University and by the Joint Research Center of the European Commission.