[en] This paper is concerned with modeling the behavior of random sums over time. Such models are particularly useful to describe the dynamics of operational losses, and to correctly estimate tail-related risk indicators. However, time-varying dependence structures make it a difficult task. To tackle these issues, we formulate a new Markov-switching generalized additive compound process combining Poisson and generalized Pareto distributions. This flexible model takes into account two important features: on the one hand, we allow all parameters of the compound loss distribution to depend on economic covariates in a flexible way. On the other hand, we allow this dependence to vary over time, via a hidden state process. A simulation study indicates that, even in the case of a short time series, this model is easily and well estimated with a standard maximum likelihood procedure. Relying on this approach, we analyze a novel dataset of 819 losses resulting from frauds at the Italian bank UniCredit. We show that our model improves the estimation of the total loss distribution over time, compared to standard alternatives. In particular, this model provides estimations of the 99.9% quantile that are never exceeded by the historical total losses, a feature
particularly desirable for banking regulators.
Research center :
Chair of Statistics and Econometrics, Faculty of Economics
Disciplines :
Quantitative methods in economics & management
Author, co-author :
Hambuckers, julien ; Université de Liège - ULiège > HEC Liège : UER > Statistique appliquée à la gestion et à l'économie
Kneib, Thomas
Langrock, Roland
Silbersdorff, Alexander
Language :
English
Title :
A Markov-switching Generalized additive model for compound Poisson processes, with applications to operational losses models
Ang, A., and Timmermann, A., Regime changes and financial markets. Ann. Rev. Financ. Econ., 2012, 4(1), 313–337.
Basel Committee on Banking Supervision BCBS, Basel II: International convergence of capital measurement and capital standards. A revised framework. Technical Report 251, Basel, Switzerland, 2004.
Beirlant, J., Goegebeur, Y., Teugels, J., Segers, J., De Waal, D., and Ferro, C., Statistics of Extremes: Theory and Applications, 2005 (Wiley: Chichester).
Bekaert, G., and Hoerova, M., The VIX, the variance premium and stock market volatility. J. Econom., 2014, 183(2), 181–192.
Chapelle, A., Crama, Y., Hübner, G., and Peters, J.-P., Practical methods for measuring and managing operational risk in the financial sector: A clinical study. J. Bank. Finance, 2008, 32(6), 1049–1061.
Chavez-Demoulin, V., Embrechts, P., and Sardy, S., Extreme-quantile tracking for financial time series. J. Econom., 2014, 181(1), 44–52.
Chavez-Demoulin, V., Embrechts, P., and Hofert, M., An extreme value approach for modeling operational risk losses depending on covariates. J. Risk Insur., 2016, 83(3), 735–776.
Chernobai, A., Jorion, P., and Yu, F., The derminants of operational risk in U.S. financial institutions. J. Financ. Quant. Anal., 2011, 46(8), 1683–1725.
Cope, E., Piche, M., and Walter, J., Macroenvironmental determinants of operational loss severity. J. Bank. Finance, 2012, 36(5), 1362–1380.
Dionne, G., and Saissi Hassani, S., Hidden markov regimes in operational loss data: Application to the recent financial crisis. J. Oper. Risk, 2017, 12(1), 23–51.
Dutta, K., and Perry, J., A tale of tails: An empirical analysis of loss distribution models for estimating operational risk capital. Res. Rev., 2006, Jul-Dec(6), 11–14.
Eilers, P., and Marx, B.D., Flexible smoothing with B-splines and penalties. Stat. Sci., 1996, 11(2), 89–102.
Embrechts, P., Klüppelberg, C., and Mikosch, T., Modelling Extremal Events for Insurance and Finance, 1997 (Springer: Berlin).
Fahrmeir, L., Kneib, T., Lang, S., and Marx, B., Regression: Models, Methods and Applications, 2013 (Springer: Berlin).
Fletcher, R., Practical Methods of Optimization, 1987 (Wiley: Chicago, IL).
Goodwin, T., Business-cycle analysis model Markov-switching. J. Bus. Econ. Stat., 1993, 11(3), 331–339.
Gray, R., Flexible methods for analyzing survival data using splines, with applications to breast cancer prognosis. J. Am. Stat. Assoc., 1992, 87(420), 942–951.
Guillou, A., Loisel, S., and Stupfler, G., Estimation of the parameters of a Markov-modulated loss process in insurance. Insur. Math. Econ., 2013, 53(2), 388–404.
Guillou, A., Loisel, S., and Stupfler, G., Estimating the parameters of a seasonal Markov-modulated Poisson process. Stat. Methodol., 2015, 26, 103–123.
Hamilton, J., A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica, 1989, 57(2), 357–384.
Hurvich, C., Simonoff, J., and Tsai, C., Information criterion smoothing parameter selection in nonparametric regression using an improved Akaike information criterion. J. R. Stat. Soc. Ser. B Stat. Methodol., 2012, 60(2), 271–293.
King, R., and Langrock, R., Semi-Markov Arnason-Schwarz models. Biometrics, 2016, 72(2), 619–628.
Laeven, L., and Levine, R., Is there a diversification discount in financial conglomerates? J. Financ. Econ., 2007, 85, 331–367.
Lamoureux, C., and Lastrapes, W., Persistence in variance, structural change, and the GARCH model. J. Bus. Econ. Stat., 1990, 8(2), 225–234.
Langrock, R., Flexible latent-state modelling of old faithful’s eruption inter-arrival times in 2009. Aust. New Zealand J. Stat., 2012, 54(3), 261–279.
Langrock, R., Kneib, T., Glennie, R., and Michelot, T., Markov-switching generalized additive models. Stat. Comput., 2017, 27(1), 259–270.
MacDonald, I.L., Numerical maximisation of the likelihood: A neglected alternative to EM? Int. Stat. Rev., 2014, 82(2), 296–308.
Marra, G., and Wood, S., Practical variable selection for generalized additive models. Comput. Stat. Data Anal., 2011, 55(7), 2372–2387.
Mayr, A., Fenske, N., Hofner, B., and Kneib, T., Generalized additive models for location, scale and shape for high dimensional data a flexible approach based on boosting. J. R. Stat. Soc. Ser. C Appl. Stat., 2012, 61(3), 403–427.
Mendoza, E., and Quadrini, V., Financial globalization, financial crises and contagion. J. Monet. Econ., 2010, 57(1), 24–39.
Moosa, I., Operational risk as a function of the state of the economy. Econ. Model., 2011, 28(5), 2137–2142.
Moscadelli, M., The modelling of operational risk: Experience with the analysis collected by the Basel Committee. Technical Report, Roma, 2004.
Neslehova, J., Embrechts, P., and Chavez-Demoulin, V., Infinite-mean models and the LDA for operational risk. J. Oper. Risk, 2006, 1(1), 3–25.
Nocedal, J., and Wright, S.J., Numerical optimization, 2nd ed., Springer Series in Operations Research, 2006 (Springer Verlag: New York).
Pesaran, M., Pettenuzzo, D., and Timmermann, A., Forecasting time series subject to multiple structural breaks. Rev. Econ. Stud., 2006, 73(4), 1057–1084.
Pohle, J., Langrock, R., van Beest, F.M., and Nabe-Nielsen, J., Selecting the number of states in hidden Markov models pragmatic solutions illustrated using animal movement. J. Agric. Biol. Environ. Stat., 2017, 22(3), 270–293.
Povel, P., Singh, R., and Winton, A., Booms, busts, and fraud. Rev. Financ. Stud., 2007, 20(4), 1219–1254.
Rigby, R., and Stasinopoulos, D., Generalized additive models for location, scale and shape. J. R. Stat. Soc. Ser. C Appl. Stat., 2005, 54(3), 507–554.
Serinaldi, F., Distributional modeling and short-term forecasting of electricity prices by generalized additive models for location, scale and shape. Energy Econ., 2011, 33(6), 1216–1226.
Sims, C., and Zha, T., Were there regime switches in U.S. monetary policy? Am. Econ. Rev., 2006, 96(1), 54–81.
Soprano, A., Crielaard, B., Piacenza, F., and Ruspantini, D., Measuring Operational and Reputational Risk: A Practitioner’s Approach, 2009 (Wiley: Chichester).
Stasinopoulos, D., and Rigby, R., Generalized Additive Models for Location Scale and Shape (GAMLSS) in R. J. Stat. Softw., 2007, 23(7), 1–46.
Wang, T., and Hsu, C., Board composition and operational risk events of financial institutions. J. Bank. Finance, 2013, 37(6), 2042–2051.
Zucchini, W., MacDonald, I., and Langrock, R., Hidden Markov Models for Time Series: An Introduction Using R, 2nd ed., 2016 (Chapman & Hall: Boca Raton, FL).