82 P. Cerchiello, M. Iannario, and D. Piccolo
Men without serious injuriesMen with serious injuries
Women without serious injuries
Women with serious injuriesExpected ratingWorking years2.8 3.2 3.6 4.0 4.40 4 8 12 16 20 24 28 32Fig. 2.Expected score as a function of working years, given gender and serious injurysuitable and flexible tool for examining and quantifying the change of response over
one or more categorical and/or continuous covariates, and they provide a deeper in-
sight into these kinds of dataset. They also allowed the summary of much information
and some interesting evaluation of specific points investigated when covariates are
both absent and present.
Moreover, we stress that the proposed model is a manifold target approach: in fact,
it can be profitably applied to a variety of fields, ranging from credit and operational
risk [3] to reputational and churn risk. Finally, it represents a convincing tool to exploit
opinions expressed by field experts.
Acknowledgement.Methodological issues and related software are partially supported by the
2006 PRIN-MIUR research project: “Stima e verifica di modelli statistici per l’analisi della
soddisfazione degli studenti universitari”.
References
- Balirano, G., Corduas, M.: Detecting semiotically expressed humor in diasporic tv pro-
ductions. Humor: Int. J. Humor Res. 3, 227–251 (2008) - Bonafede, E.C., Cerchiello, P., Giudici, P.: A proposal for the fuzzification of causal vari-
ables, in operational risk management. Quaderni di Dipartimento di Statistica 44 (2007) - Bonafede, E.C., Cerchiello, P., Giudici, P.: Statistical models for Business Continuity
Management. J. Oper. Risk Man. 2, 79–96 (2007) - Bonafede, E.C., Giudici, P.: Bayesian networks for enterprise risk assessment. Phys. A
Stat. Mech. Appl. 382, 22–28 (2007) - Bromiley, P., Curley, S.P.: Individual differences in risk taking. In: Yates, F. (ed.) Risk-
Taking Behavior. pp. 87–132. Wiley, New York (1992)