A GENERALIZED ADDITIVE MODEL FOR BINARY RARE
EVENTS DATA: AN APPLICATION TO CREDIT DEFAULTS

Silvia Angela Osmetti; Raffaella Calabrese

A GENERALIZED ADDITIVE MODEL FOR BINARY RARE EVENTS DATA: AN APPLICATION TO CREDIT DEFAULTS

Silvia Angela Osmetti, Raffaella Calabrese

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We aim at proposing a new model for binary rare events, i.e. binary dependent variable with a very small number of ones. We extend the Generalized Extreme Value (GEV) regression model (Calabrese and Osmetti, 2011) to a Generalized 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). We obtain that the GEVA model shows a high predictive accuracy to identify the rare event.

Original language	English
Title of host publication	Analysis and Modeling of Complex Data in Behavioural and Social Sciences In Book of short papers JCS
Pages	1-4
Number of pages	4
Publication status	Published - 2012
Event	JCS - CLADAG 2012 - Anacapri Duration: 3 Sept 2012 → 4 Sept 2012

Conference

Conference	JCS - CLADAG 2012
City	Anacapri
Period	3/9/12 → 4/9/12

Keywords

CREDIT DEFAULT
GENERALIZED ADDITIVE MODEL

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Cite this

@inproceedings{e4df94193a4b4d71a1f45ed46a44cf61,

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 dependent variable with a very small number of ones. We extend the Generalized Extreme Value (GEV) regression model (Calabrese and Osmetti, 2011) to a Generalized 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). We obtain that the GEVA model shows a high predictive accuracy to identify the rare event.",

keywords = "CREDIT DEFAULT, GENERALIZED ADDITIVE MODEL, CREDIT DEFAULT, GENERALIZED ADDITIVE MODEL",

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

year = "2012",

language = "English",

isbn = "978-88-6129-916-0",

pages = "1--4",

booktitle = "Analysis and Modeling of Complex Data in Behavioural and Social Sciences In Book of short papers JCS",

note = "JCS - CLADAG 2012 ; Conference date: 03-09-2012 Through 04-09-2012",

}

TY - GEN

T1 - A GENERALIZED ADDITIVE MODEL FOR BINARY RARE EVENTS DATA: AN APPLICATION TO CREDIT DEFAULTS

AU - Osmetti, Silvia Angela

AU - Calabrese, Raffaella

PY - 2012

Y1 - 2012

N2 - We aim at proposing a new model for binary rare events, i.e. binary dependent variable with a very small number of ones. We extend the Generalized Extreme Value (GEV) regression model (Calabrese and Osmetti, 2011) to a Generalized 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). We obtain that the GEVA model shows a high predictive accuracy to identify the rare event.

AB - We aim at proposing a new model for binary rare events, i.e. binary dependent variable with a very small number of ones. We extend the Generalized Extreme Value (GEV) regression model (Calabrese and Osmetti, 2011) to a Generalized 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). We obtain that the GEVA model shows a high predictive accuracy to identify the rare event.

KW - CREDIT DEFAULT

KW - GENERALIZED ADDITIVE MODEL

KW - CREDIT DEFAULT

KW - GENERALIZED ADDITIVE MODEL

UR - http://hdl.handle.net/10807/31688

M3 - Conference contribution

SN - 978-88-6129-916-0

SP - 1

EP - 4

BT - Analysis and Modeling of Complex Data in Behavioural and Social Sciences In Book of short papers JCS

T2 - JCS - CLADAG 2012

Y2 - 3 September 2012 through 4 September 2012

ER -