A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults.

Raffaella Calabrese; Silvia Angela Osmetti

doi:10.1007/978-3-319-06692-9_9

A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults.

Raffaella Calabrese, Silvia Angela Osmetti

Risultato della ricerca: Contributo in libro › Chapter

1 Citazioni (Scopus)

Abstract

We aim at proposing a new model for binary rare events, i.e. binary depen- dent variable with a very small number of ones.We extend the Generalized Extreme Value (GEV) regression model proposed by Calabrese and Osmetti [5] to a Gener- alized Additive Model (GAM). We suggest to consider the quantile function of the GEV distribution as a link function in a GAM, so we propose the Generalized Extreme Value Additive (GEVA) model. In order to estimate the GEVA model, a modified version of the local scoring algorithm of GAM is proposed. Finally, to model default probability, we apply our proposal to empirical data on Italian Small and Medium Enterprises (SMEs). The results show that the GEVA model has a higher predictive accuracy to identify the rare event than the logistic additive model.

Lingua originale	English
Titolo della pubblicazione ospite	Analysis and Modeling of Complex Data in Behavioural and Social Sciences
Editor	D Vicari, A Okada, G Ragozini, C Weihs
Pagine	73-81
Numero di pagine	9
DOI	https://doi.org/10.1007/978-3-319-06692-9_9
Stato di pubblicazione	Pubblicato - 2014

Serie di pubblicazioni

Nome	STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION

Keywords

credit defaults
generalized additive model
generalized extreme value distribution
local scoring algorithm

OSS delle Nazioni Unite

Questo processo contribuisce al raggiungimento dei seguenti obiettivi di sviluppo sostenibile

Accesso al documento

10.1007/978-3-319-06692-9_9

Altri file e collegamenti

Link a IRIS PubliCatt

Cita questo

Calabrese, R., & Osmetti, S. A. (2014). A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults. In D. Vicari, A. Okada, G. Ragozini, & C. Weihs (Edd.), Analysis and Modeling of Complex Data in Behavioural and Social Sciences (pagg. 73-81). (STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION). https://doi.org/10.1007/978-3-319-06692-9_9

@inbook{2b2b6dd909a24fc29fcce026398d044c,

title = "A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults.",

abstract = "We aim at proposing a new model for binary rare events, i.e. binary depen- dent variable with a very small number of ones.We extend the Generalized Extreme Value (GEV) regression model proposed by Calabrese and Osmetti [5] to a Gener- alized Additive Model (GAM). We suggest to consider the quantile function of the GEV distribution as a link function in a GAM, so we propose the Generalized Extreme Value Additive (GEVA) model. In order to estimate the GEVA model, a modified version of the local scoring algorithm of GAM is proposed. Finally, to model default probability, we apply our proposal to empirical data on Italian Small and Medium Enterprises (SMEs). The results show that the GEVA model has a higher predictive accuracy to identify the rare event than the logistic additive model.",

keywords = "credit defaults, generalized additive model, generalized extreme value distribution, local scoring algorithm, credit defaults, generalized additive model, generalized extreme value distribution, local scoring algorithm",

author = "Raffaella Calabrese and Osmetti, {Silvia Angela}",

year = "2014",

doi = "10.1007/978-3-319-06692-9_9",

language = "English",

isbn = "978-3-319-06691-2",

series = "STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION",

pages = "73--81",

editor = "D Vicari and A Okada and G Ragozini and C Weihs",

booktitle = "Analysis and Modeling of Complex Data in Behavioural and Social Sciences",

}

Calabrese, R & Osmetti, SA 2014, A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults. in D Vicari, A Okada, G Ragozini & C Weihs (edd.), Analysis and Modeling of Complex Data in Behavioural and Social Sciences. STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION, pagg. 73-81. https://doi.org/10.1007/978-3-319-06692-9_9

A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults. / Calabrese, Raffaella; Osmetti, Silvia Angela.
Analysis and Modeling of Complex Data in Behavioural and Social Sciences. a cura di / D Vicari; A Okada; G Ragozini; C Weihs. 2014. pag. 73-81 (STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION).

Risultato della ricerca: Contributo in libro › Chapter

TY - CHAP

T1 - A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults.

AU - Calabrese, Raffaella

AU - Osmetti, Silvia Angela

PY - 2014

Y1 - 2014

N2 - We aim at proposing a new model for binary rare events, i.e. binary depen- dent variable with a very small number of ones.We extend the Generalized Extreme Value (GEV) regression model proposed by Calabrese and Osmetti [5] to a Gener- alized Additive Model (GAM). We suggest to consider the quantile function of the GEV distribution as a link function in a GAM, so we propose the Generalized Extreme Value Additive (GEVA) model. In order to estimate the GEVA model, a modified version of the local scoring algorithm of GAM is proposed. Finally, to model default probability, we apply our proposal to empirical data on Italian Small and Medium Enterprises (SMEs). The results show that the GEVA model has a higher predictive accuracy to identify the rare event than the logistic additive model.

AB - We aim at proposing a new model for binary rare events, i.e. binary depen- dent variable with a very small number of ones.We extend the Generalized Extreme Value (GEV) regression model proposed by Calabrese and Osmetti [5] to a Gener- alized Additive Model (GAM). We suggest to consider the quantile function of the GEV distribution as a link function in a GAM, so we propose the Generalized Extreme Value Additive (GEVA) model. In order to estimate the GEVA model, a modified version of the local scoring algorithm of GAM is proposed. Finally, to model default probability, we apply our proposal to empirical data on Italian Small and Medium Enterprises (SMEs). The results show that the GEVA model has a higher predictive accuracy to identify the rare event than the logistic additive model.

KW - credit defaults

KW - generalized additive model

KW - generalized extreme value distribution

KW - local scoring algorithm

KW - credit defaults

KW - generalized additive model

KW - generalized extreme value distribution

KW - local scoring algorithm

UR - http://hdl.handle.net/10807/56463

U2 - 10.1007/978-3-319-06692-9_9

DO - 10.1007/978-3-319-06692-9_9

M3 - Chapter

SN - 978-3-319-06691-2

T3 - STUDIES IN CLASSIFICATION, DATA ANALYSIS, AND KNOWLEDGE ORGANIZATION

SP - 73

EP - 81

BT - Analysis and Modeling of Complex Data in Behavioural and Social Sciences

A2 - Vicari, D

A2 - Okada, A

A2 - Ragozini, G

A2 - Weihs, C

ER -

A Generalized Additive Model for Binary Rare Events Data: an Application to Credit Defaults.

Abstract

Serie di pubblicazioni

Keywords

OSS delle Nazioni Unite

Accesso al documento

Altri file e collegamenti

Fingerprint

Cita questo