Estimating binary spatial autoregressive models for rare events

Raffaella Calabrese; Johan Elkink

doi:10.1108/S0731-905320160000037012

Estimating binary spatial autoregressive models for rare events

Research output: Chapter in Book/Report/Conference proceeding › Chapter (peer-reviewed) › peer-review

Abstract / Description of output

The most used spatial regression models for binary dependent variable consider a symmetric link function, such as the logistic or the probit models. When the dependent variable represents a rare event, a symmetric link function can underestimate the probability that the rare event occurs. Following Calabrese and Osmetti (2013), we suggest the quantile function
of the Generalized Extreme Value (GEV) distribution as link function in a spatial generalized linear model and we call this model the Spatial GEV (SGEV) regression model. To estimate the parameters of such model, a modified version of the Gibbs sampling method of Wang and Dey (2010) is proposed. We analyze the performance of our model by Monte Carlo simulationsand evaluate the prediction accuracy in empirical data on state failure.

Original language	English
Title of host publication	Advances in Econometrics
Publisher	Emerald Publishing
Pages	145-166
Volume	37
ISBN (Electronic)	9781785609855
ISBN (Print)	9781785609862
DOIs	https://doi.org/10.1108/S0731-905320160000037012
Publication status	Published - 2016

Publication series

Name	Advances in Econometrics
Volume	37
ISSN (Print)	0731-9053

Access to Document

10.1108/S0731-905320160000037012

paper
'This article is © Emerald Group Publishing and permission has been granted for this version to appear here http://www.research.ed.ac.uk/portal/en/publications/estimating-binary-spatial-autoregressive-models-for-rare-events(40014934-d8db-41a0-8d78-59518b52d892).html Emerald does not grant permission for this article to be further copied/distributed or hosted elsewhere without the express permission from Emerald Group Publishing Limited.'
Accepted author manuscript, 185 KB

Cite this

@inbook{40014934d8db41a08d7859518b52d892,

title = "Estimating binary spatial autoregressive models for rare events",

abstract = "The most used spatial regression models for binary dependent variable consider a symmetric link function, such as the logistic or the probit models. When the dependent variable represents a rare event, a symmetric link function can underestimate the probability that the rare event occurs. Following Calabrese and Osmetti (2013), we suggest the quantile functionof the Generalized Extreme Value (GEV) distribution as link function in a spatial generalized linear model and we call this model the Spatial GEV (SGEV) regression model. To estimate the parameters of such model, a modified version of the Gibbs sampling method of Wang and Dey (2010) is proposed. We analyze the performance of our model by Monte Carlo simulationsand evaluate the prediction accuracy in empirical data on state failure.",

author = "Raffaella Calabrese and Johan Elkink",

year = "2016",

doi = "10.1108/S0731-905320160000037012",

language = "English",

isbn = "9781785609862",

volume = "37",

series = "Advances in Econometrics",

publisher = "Emerald Publishing",

pages = "145--166",

booktitle = "Advances in Econometrics",

}

TY - CHAP

T1 - Estimating binary spatial autoregressive models for rare events

AU - Calabrese, Raffaella

AU - Elkink, Johan

PY - 2016

Y1 - 2016

N2 - The most used spatial regression models for binary dependent variable consider a symmetric link function, such as the logistic or the probit models. When the dependent variable represents a rare event, a symmetric link function can underestimate the probability that the rare event occurs. Following Calabrese and Osmetti (2013), we suggest the quantile functionof the Generalized Extreme Value (GEV) distribution as link function in a spatial generalized linear model and we call this model the Spatial GEV (SGEV) regression model. To estimate the parameters of such model, a modified version of the Gibbs sampling method of Wang and Dey (2010) is proposed. We analyze the performance of our model by Monte Carlo simulationsand evaluate the prediction accuracy in empirical data on state failure.

AB - The most used spatial regression models for binary dependent variable consider a symmetric link function, such as the logistic or the probit models. When the dependent variable represents a rare event, a symmetric link function can underestimate the probability that the rare event occurs. Following Calabrese and Osmetti (2013), we suggest the quantile functionof the Generalized Extreme Value (GEV) distribution as link function in a spatial generalized linear model and we call this model the Spatial GEV (SGEV) regression model. To estimate the parameters of such model, a modified version of the Gibbs sampling method of Wang and Dey (2010) is proposed. We analyze the performance of our model by Monte Carlo simulationsand evaluate the prediction accuracy in empirical data on state failure.

U2 - 10.1108/S0731-905320160000037012

DO - 10.1108/S0731-905320160000037012

M3 - Chapter (peer-reviewed)

SN - 9781785609862

VL - 37

T3 - Advances in Econometrics

SP - 145

EP - 166

BT - Advances in Econometrics

PB - Emerald Publishing

ER -

Estimating binary spatial autoregressive models for rare events

Abstract / Description of output

Publication series

Access to Document

Fingerprint

Cite this