LES PREVISIONS STATIQUES ET DYNAMIQUES DES VALEURS A RISQUE (VaR) DES ACTIONS DE BANQUE : LE MODELE DE DEPASSEMENTS DE SEUIL (POT) ET LES MODELES DE SCORE AUTOREGRESSIFS GENERALISES (GAS)
Author :
Abstract
Le calcul de la valeur à risque des actifs est un moyen fréquemment utilisé pour mesurer le risque financier. Les Valeurs à Risque peuvent être évaluées et interprétées dans le cadre de la théorie des valeurs extrêmes en raison des valeurs excessives dues aux incertitudes des rendements. Dans cette étude, les prévisions statiques et dynamiques de la VaR des actions de cinq banques opérant sur les marchés des pays développés et en développement sont établies à l'aide des modèles POT et GAS, respectivement. Le nombre des valeurs extrêmes requises pour l'estimation du modèle POT a été calculé séparément en utilisant les méthodes de racine carrée et 10% du nombre total de données, qui sont deux méthodes pratiques dans la littérature. Les résultats ont été interprétés séparément. De plus, les modèles GAS basés sur la distribution normale et celle de student - t et de student - t asymétrique ont été estimés et comparés les uns aux autres.
Keywords
Abstract
The calculation of the value at risk of assets is a frequently used means of measuring financial risk. Risk Values may be valued and interpreted within the framework of extreme value theory because of excessive values due to uncertainties in returns. In this study, the static and dynamic VaR forecasts of the stocks of five banks operating in developed and developing country markets are determined using the POT and GAS models, respectively. The number of extreme values required for the estimation of the POT model was calculated separately using square root and 10% of the total number of data, which are two practical methods in the literature. The results were interpreted separately. In addition, GAS models based on normal, student-t and asymmetric student-t distributions were estimated and compared to each other.
Keywords
- Andjelic, G., Milosev, I. et Djakovic, V. (2010). Extreme Value Theory in Emerging Markets. Economic Annals, 55 / 185, 63 – 105.
- Ardia, D., Boudt, K. et Catania, L. (2016). Value at Risk Prediction in R with the GAS Package. SSRN Electronic Journal, 10.2139/ssrn.2871444.
- Arık, A., Bulut, B. et Sucu, M. (2013). Finansal Risklerin Uç Değer Kuramı ile Ölçülmesi. Bilim ve Teknoloji Dergisi, 14 / 2, 119 – 134.
- Avşarlıgil, N., Demir, Y. et Doğru, E. (2015). Riske Maruz Değer Ölçüm Yöntemleri Aracılığıyla BIST’te İşlem Gören Spor Kulüpleri Üzerine Bir Uygulama. Eskişehir Osmangazi Üniversitesi Sosyal Bilimler Dergisi, 16 / 1, 81 – 107.
- Blasques, F., Koopman, S.J. et Lucas, A. (2014). Maximum Likelihood Estimation for correctly Specified Generalized Autoregressive Score Models : Feedback Effects, Contraction Conditions and Asymptotic Properties. Tibergen Institute, 14 – 074 / III, 1 – 29.
- Blasques, F. et al. (2015). In – Sample Confidence Bands and Out – of – Sample Forecast Bands for Time - Varying Parameters in Observation Driven Models. Tinbergen Institute, 15 – 083 / III, 1 – 32.
- Bostancı, A. et Korkmaz, T. (2014). Bankaların Sermaye Yeterliliği Oranı Açısından Riske Maruz Değer Hesaplama Yöntemlerinin Karşılaştırılması. Business and Economics Research Journal, 5 / 3, 15 – 41.
- Creal, D., Koopman, S.J. et Lucas, A. (2013). Generalized Autoregressive Score Models with Applications. Journal of Applied Econometrics, 28, 777 – 795.
- Çelik, M. et Kaya, M.F. (2010). Uç Değerler Yöntemi ile Riske Maruz Değer’in Tahmini ve İstanbul Menkul Kıymetler Borsası Üzerine Bir Uygulama. Bankacılık ve Sigortacılık Araştırmaları Dergisi, 1 / 1, 19 – 32.
- Çifter, A., Özün, A. et Yılmazer, S. (2007). Beklenen Kuyruk Kaybı ve Genelleştirlmiş Pareto Dağılımı ile Riske Maruz Değer Öngörüsü. Bankacılar Dergisi, 60, 3 – 16.
- Dumouchel, W.H. (1983). Estimating the Stable Index in Order to Measure Tail Thickness : A Critique. The Annals of Statistics, 11 / 4, 1019 – 1031.
- Engle, R.F. et Manganelli, S. (2004). CAViaR : Conditional Autoregressive Value at Risk by Regression Quantiles. Journal of Business & Economic Statistics, 22 / 4, 367 – 381.
- Evci, S. et Kandır, S.Y. (2015). Endüstriyel Metal Piyasasında Piyasa Riskinin Ölçülmesi : Riske Maruz Değer (VaR) Yöntemi ile Bir Uygulama. Niğde Üniversitesi İ.İ.B.F. Dergisi, 8 / 1, 157 – 170.
- Ferreira, A., Haan, L et Peng, L. (2003). On Optimising the Estimation of High Quantiles of a Probability Distribution. Statistics : A Journal of Theoretical and Applied Statistics, 37 / 5, 401 – 434.
- Hosking, J.R.M. et Wallis, J.R. (1987). Parameter and Quantile Estimation for the Generalized Pareto Distribution. Technometrics, 29 / 3, 339 -349.
- Jorion, P. (2007). Value at Risk. 3rd Ed. New York : The McGraw – Hill.
- Koopman, S.J., Lucas, A. et Scharth, M. (2012). Predicting Time – Varying Parameters with Parameter – Driven and Observation - Driven Models. Tinbergen Institute, 12 – 020 / 4, 1 – 37.
- McNeil, A.J., Frey, R. et Embrechts, P. (2005). Quantitative Risk Management : Concepts, Techniques and Tools. New Jersey : Princeton University Press.
- Nortey, E., Asare, K. et Mettle, F.O. (2015). Extreme Value Modelling of Ghana Stock Exchange Index. SpringerPlus, 4 / 696, 1 – 17.
- Pickands, J. (1975). Statistical Inference Using Extreme Order Statistics. The Annals of Statis- tics, 3 / 1, 119 – 131.
- Rivera, G.G., Lee, T.H. et Mishra, S. (2004). Forecasting Volatility : A Reality Check Based on Option Pricing, Utility Function, Value – at - Risk and Predictive Likelihood. International Journal of Forecasting, 20, 629 – 645.
- Scarrott, C. et MacDonald, A. (2012). A Review of Extreme Value Threshold Estimation and Uncertainty Quantification. REVSTAT, 10 / 1, 33 – 60.
- Totic, S. (2015). A Conditional Extreme Value Theory Approach in Value – At - Risk Fore- casting : Evidence from Southeastern Europe and USA market. Industrija, 43 / 4, 7 –
- Ural, M. et Adakale, T. (2009). Beklenen Kayıp Yöntemi ile Riske Maruz Değer Analizi. Akdeniz İ.İ.B.F. Dergisi, 17, 23 – 39.
