In this paper, a novel approach for model comparison is presented. The standard linear regression model framework is considered and extended through the employment of the Lorenz zonoid tool. By means of the Lorenz zonoid, which in the univariate case corresponds to the classical Gini coefficient, a twofold aim is achieved. On the one hand the contribution of any additional independent variable introduced into the linear model is determined and, on the other hand, a relative measure of the variability characterising the underlying response variable is pro- posed. The results is a Lorenz zonoids-based dependence measure which mimics the partial correlation coefficient by simplifying the interpretation, especially when dealing with phenomena affected by high variability.
Lorenz zonoid measures to compare predictive accuracy / P. Giudici, E. Raffinetti. ((Intervento presentato al convegno IFABS Conference : Reinventing banking and sustainable finance tenutosi a Angers nel 2019.
Lorenz zonoid measures to compare predictive accuracy
E. Raffinetti
2019
Abstract
In this paper, a novel approach for model comparison is presented. The standard linear regression model framework is considered and extended through the employment of the Lorenz zonoid tool. By means of the Lorenz zonoid, which in the univariate case corresponds to the classical Gini coefficient, a twofold aim is achieved. On the one hand the contribution of any additional independent variable introduced into the linear model is determined and, on the other hand, a relative measure of the variability characterising the underlying response variable is pro- posed. The results is a Lorenz zonoids-based dependence measure which mimics the partial correlation coefficient by simplifying the interpretation, especially when dealing with phenomena affected by high variability.File | Dimensione | Formato | |
---|---|---|---|
Paper_IFABS_2019.pdf
accesso aperto
Tipologia:
Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Dimensione
579.96 kB
Formato
Adobe PDF
|
579.96 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.