Estimating Binary Spatial Autoregressive Models for Rare Events

This paper proposes a new statistical estimator, to be applied to the prediction of state failures. State failures are typically conceptualised in a binary fashion—a state fails or it does not—and are rare events. Furthermore, state failures are not geographically independent events. The failure of one state can be expected to have an impact on the stability and peace in neighboring states, increasing the probability of state failures in geographical contiguous regions. Currently there is mixed evidence of such diffusion of state failure taking place Iqbal and Starr (2008). This paper proposes a new estimator to be used for estimates of the spatial interdependence among state failures and focuses on the ability to predict state failures in an international context. The most used spatial regression models for binary dependent variable consider a symmetric link function. When the dependent variable represents a rare event, a symmetric link function is not coherent. Following Calabrese and Osmetti (2013), we suggest the quantile function of the Generalized Extreme Value (GEV) distribution as link function in a spatial generalised 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 LeSage (2000) and Wang and Dey (2010) is proposed. We analyse the performance of our model by Monte Carlo simulations and evaluating the prediction quality in empirical data on state failure.

[1]  R. Pace,et al.  Modeling Spatially Interdependent Mortgage Decisions , 2011 .

[2]  Kristian Skrede Gleditsch,et al.  Mapping and Measuring Country Shapes , 2010 .

[3]  Max Welling,et al.  Austerity in MCMC Land: Cutting the Metropolis-Hastings Budget , 2013, ICML 2014.

[4]  Christoforos Anagnostopoulos,et al.  When is the area under the receiver operating characteristic curve an appropriate measure of classifier performance? , 2013, Pattern Recognit. Lett..

[5]  C. Schär,et al.  A PRECIPITATION CLIMATOLOGY OF THE ALPS FROM HIGH-RESOLUTION RAIN-GAUGE OBSERVATIONS , 1998 .

[6]  Marno Verbeek,et al.  A Guide to Modern Econometrics , 2000 .

[7]  J. LeSage.  Bayesian Estimation of Limited Dependent Variable Spatial Autoregressive Models , 2010 .

[8]  Silvia Angela Osmetti,et al.  Modelling small and medium enterprise loan defaults as rare events: the generalized extreme value regression model , 2013 .

[9]  Gary King,et al.  Logistic Regression in Rare Events Data , 2001, Political Analysis.

[10]  Eric R. Ziegel,et al.  Generalized Linear Models , 2002, Technometrics.

[11]  Roger Bivand,et al.  A review of spatial statistical techniques for location studies , 1998 .

[12]  D. McMillen.  PROBIT WITH SPATIAL AUTOCORRELATION , 1992 .

[13]  Paolo Giudici,et al.  Measuring bank contagion in Europe using binary spatial regression models , 2017, J. Oper. Res. Soc..

[14]  Gary M. Weiss.  Mining with rarity: a unifying framework , 2004, SKDD.

[15]  David J. Hand,et al.  Measuring classifier performance: a coherent alternative to the area under the ROC curve , 2009, Machine Learning.

[16]  J. Goldstone,et al.  A Global Model for Forecasting Political Instability , 2010 .

[17]  Thomas H. Klier,et al.  Clustering of Auto Supplier Plants in the United States , 2008 .

[18]  Ryan P. Adams,et al.  Firefly Monte Carlo: Exact MCMC with Subsets of Data , 2014, UAI.

[19]  Joris Pinkse,et al.  Contracting in space: An application of spatial statistics to discrete-choice models , 1998 .

[20]  J. Hüsler,et al.  Laws of Small Numbers: Extremes and Rare Events , 1994 .

[21]  Alan Agresti,et al.  Categorical Data Analysis , 2003 .

[22]  Christoforos Anagnostopoulos,et al.  A better Beta for the H measure of classification performance , 2012, Pattern Recognit. Lett..

[23]  Kristian Skrede Gleditsch,et al.  Mapping and Measuring Country Shapes: The cshapes Package , 2010 .

[24]  A. Scott,et al.  Fitting Logistic Models Under Case‐Control or Choice Based Sampling , 1986 .

[25]  Johan A. Elkink.  The International Diffusion of Democracy , 2011 .

[26]  Álvaro A. Novo.  Contagious Currency Crisis: A Spatial Probit Approach , 2003 .

[27]  Paul W Dickman,et al.  Estimating and modeling the cure fraction in population-based cancer survival analysis. , 2007, Biostatistics.

[28]  S. Roberts.  EXTREME VALUE STATISTICS FOR NOVELTY DETECTION IN BIOMEDICAL DATA PROCESSING , 2000 .

[29]  S. Nadarajah,et al.  Extreme Value Distributions: Theory and Applications , 2000 .

[30]  Raffaella Calabrese,et al.  Estimators of Binary Spatial Autoregressive Models: A Monte Carlo Study , 2013, 1302.3414.

[31]  Michael D. Ward,et al.  Forecasting is difficult, especially about the future , 2013 .

[32]  DJ Hand,et al.  Performance criteria for plastic card fraud detection tools , 2008, J. Oper. Res. Soc..

[33]  Michael J. Donahoo,et al.  Under the Hood , 2009 .

[34]  R. Calabrese,et al.  Optimal cut-off for rare events and unbalanced misclassification costs , 2014 .

[35]  Diana Barro,et al.  Credit contagion in a network of firms with spatial interaction , 2010, Eur. J. Oper. Res..

[36]  Steven R. Lerman,et al.  The Estimation of Choice Probabilities from Choice Based Samples , 1977 .

[37]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[38]  Harvey E. Starr.  Civil War: Spatiality, Contagion, and Diffusion , 2009 .

[39]  Wim P. M. Vijverberg,et al.  Probit in a Spatial Context: A Monte Carlo Analysis , 2004 .

[40]  Richard L. Smith.  Maximum likelihood estimation in a class of nonregular cases , 1985 .

[41]  Cláudia Neves,et al.  Extreme Value Distributions , 2011, International Encyclopedia of Statistical Science.

[42]  Harvey E. Starr,et al.  Bad Neighbors: Failed States and Their Consequences , 2008 .

[43]  D. McMillen.  Spatial Effects in Probit Models: A Monte Carlo Investigation , 1995 .

[44]  Eddie Kohler,et al.  Accelerating MCMC via Parallel Predictive Prefetching , 2014, UAI.

[45]  Kelly E. Murray,et al.  Under the Hood , 1996, J. Object Oriented Program..

[46]  Dipak K. Dey,et al.  Generalized extreme value regression for binary response data: An application to B2B electronic payments system adoption , 2011, 1101.1373.

[47]  Nina S. N. Lam,et al.  New Orleans business recovery in the aftermath of Hurricane Katrina , 2011 .

[48]  Luc Anselin,et al.  Under the hood , 2002 .

[49]  Silvia Angela Osmetti,et al.  Improving Forecast of Binary Rare Events Data: A GAM-Based Approach , 2015 .

[50]  Gary King,et al.  Explaining Rare Events in International Relations , 2001, International Organization.

[51]  Luc Anselin,et al.  A Note on Small Sample Properties of Estimators in a First-Order Spatial Autoregressive Model , 1982 .

[52]  Maureen A. Pirog-Good,et al.  An analysis of youth crime and employment patterns , 1986 .

[53]  Robin C. Sickles,et al.  Banking Crises, Early Warning Models, and Efficiency , 2016 .

[54]  M. D. Ugarte,et al.  Introduction to Spatial Econometrics , 2011 .

[55]  Mark M. Fleming.  Techniques for Estimating Spatially Dependent Discrete Choice Models , 2004 .

[56]  Michael D. Ward,et al.  Diffusion and the International Context of Democratization , 2006, International Organization.