On the tail dependence in bivariate hydrological frequency analysis

Alexandre Lekina; Fateh Chebana; Taha B. M. J. Ouarda

Dependence Modeling (2015)

  • Volume: 3, Issue: 1, page 203-227, electronic only
  • ISSN: 2300-2298

Abstract

top
In Bivariate Frequency Analysis (BFA) of hydrological events, the study and quantification of the dependence between several variables of interest is commonly carried out through Pearson’s correlation (r), Kendall’s tau (τ) or Spearman’s rho (ρ). These measures provide an overall evaluation of the dependence. However, in BFA, the focus is on the extreme events which occur on the tail of the distribution. Therefore, these measures are not appropriate to quantify the dependence in the tail distribution. To quantify such a risk, in Extreme Value Analysis (EVA), a number of concepts and methods are available but are not appropriately employed in hydrological BFA. In the present paper, we study the tail dependence measures with their nonparametric estimations. In order to cover a wide range of possible cases, an application dealing with bivariate flood characteristics (peak flow, flood volume and event duration) is carried out on three gauging sites in Canada. Results show that r, τ and ρ are inadequate to quantify the extreme risk and to reflect the dependence characteristics in the tail. In addition, the upper tail dependence measure, commonly employed in hydrology, is shown not to be always appropriate especially when considered alone: it can lead to an overestimation or underestimation of the risk. Therefore, for an effective risk assessment, it is recommended to consider more than one tail dependence measure.

How to cite

top

Alexandre Lekina, Fateh Chebana, and Taha B. M. J. Ouarda. "On the tail dependence in bivariate hydrological frequency analysis." Dependence Modeling 3.1 (2015): 203-227, electronic only. <http://eudml.org/doc/275981>.

@article{AlexandreLekina2015,
abstract = {In Bivariate Frequency Analysis (BFA) of hydrological events, the study and quantification of the dependence between several variables of interest is commonly carried out through Pearson’s correlation (r), Kendall’s tau (τ) or Spearman’s rho (ρ). These measures provide an overall evaluation of the dependence. However, in BFA, the focus is on the extreme events which occur on the tail of the distribution. Therefore, these measures are not appropriate to quantify the dependence in the tail distribution. To quantify such a risk, in Extreme Value Analysis (EVA), a number of concepts and methods are available but are not appropriately employed in hydrological BFA. In the present paper, we study the tail dependence measures with their nonparametric estimations. In order to cover a wide range of possible cases, an application dealing with bivariate flood characteristics (peak flow, flood volume and event duration) is carried out on three gauging sites in Canada. Results show that r, τ and ρ are inadequate to quantify the extreme risk and to reflect the dependence characteristics in the tail. In addition, the upper tail dependence measure, commonly employed in hydrology, is shown not to be always appropriate especially when considered alone: it can lead to an overestimation or underestimation of the risk. Therefore, for an effective risk assessment, it is recommended to consider more than one tail dependence measure.},
author = {Alexandre Lekina, Fateh Chebana, Taha B. M. J. Ouarda},
journal = {Dependence Modeling},
keywords = {Asymptotic; Extreme; Copula; Bivariate distribution; Non-parametric estimation; asymptotic; extreme; copula; bivariate distribution; nonparametric estimation},
language = {eng},
number = {1},
pages = {203-227, electronic only},
title = {On the tail dependence in bivariate hydrological frequency analysis},
url = {http://eudml.org/doc/275981},
volume = {3},
year = {2015},
}

TY - JOUR
AU - Alexandre Lekina
AU - Fateh Chebana
AU - Taha B. M. J. Ouarda
TI - On the tail dependence in bivariate hydrological frequency analysis
JO - Dependence Modeling
PY - 2015
VL - 3
IS - 1
SP - 203
EP - 227, electronic only
AB - In Bivariate Frequency Analysis (BFA) of hydrological events, the study and quantification of the dependence between several variables of interest is commonly carried out through Pearson’s correlation (r), Kendall’s tau (τ) or Spearman’s rho (ρ). These measures provide an overall evaluation of the dependence. However, in BFA, the focus is on the extreme events which occur on the tail of the distribution. Therefore, these measures are not appropriate to quantify the dependence in the tail distribution. To quantify such a risk, in Extreme Value Analysis (EVA), a number of concepts and methods are available but are not appropriately employed in hydrological BFA. In the present paper, we study the tail dependence measures with their nonparametric estimations. In order to cover a wide range of possible cases, an application dealing with bivariate flood characteristics (peak flow, flood volume and event duration) is carried out on three gauging sites in Canada. Results show that r, τ and ρ are inadequate to quantify the extreme risk and to reflect the dependence characteristics in the tail. In addition, the upper tail dependence measure, commonly employed in hydrology, is shown not to be always appropriate especially when considered alone: it can lead to an overestimation or underestimation of the risk. Therefore, for an effective risk assessment, it is recommended to consider more than one tail dependence measure.
LA - eng
KW - Asymptotic; Extreme; Copula; Bivariate distribution; Non-parametric estimation; asymptotic; extreme; copula; bivariate distribution; nonparametric estimation
UR - http://eudml.org/doc/275981
ER -

References

top
  1. [1] Ané, T. and C. Kharoubi (2003). Dependence structure and risk measure. J. Business 76, 411–438. [Crossref] 
  2. [2] Bacro, J. N. (2005). Dépendance des extrêmes. Tech. rep., Institut de Mathématiques et de Modélisation de Montpellier. 
  3. [3] Beirlant, J., Y. Goegebeur, J. Teugels, and J. Segers (2004). Statistics of Extremes: Theory and Applications. John Wiley and Sons, Chichester. Zbl1070.62036
  4. [4] Ben Aissia, M.-A., F. Chebana, T. B. M. J. Ouarda, L. Roy, G. Desrochers, I. Chartier, and E. Robichaud (2012). Multivariate analysis of flood characteristics in a climate change context of thewatershed of the Baskatong reservoir, Province of Québec, Canada. Hydrol. Process. 26(1), 130–142. [Crossref] 
  5. [5] Bingham, N. H., C. M. Goldie, and J. L. Teugels (1987). Regular Variation. Cambridge University Press. Zbl0617.26001
  6. [6] Bobée, B. and F. Ashkar (1991). The Gamma Family and Derived Distributions Applied in Hydrology. Water Resources Publications, Colorado. 
  7. [7] Bouchouicha, R. (2010). Dépendance entre risques extrêmes: application aux hedge funds. Tech. Rep. 1013, Groupe d’Analyse et de Théorie Economique (GATE), Centre national de la recherche scientifique (CNRS), Université Lyon 2, Ecole Normale Supérieure. 
  8. [8] Capéraà, P., A.-L. Fougères, and C. Genest (1997). A nonparametric estimation procedure for bivariate extreme value copulas. Biometrika 84(3), 567–577. [Crossref] Zbl1058.62516
  9. [9] Chebana, F. (2013). Multivariate analysis of hydrological variables. In Encyclopedia of Environmetrics, Second Edition, A. H. El-Shaarawi and W. W. Piegorsch eds., 1676–1681. John Wiley and Sons, Chichester. 
  10. [10] Chebana, F. and T. Ouarda (2011). Depth-based multivariate descriptive statistics with hydrological applications. J. Geophys. Res. 116(D10), 1–19. [Crossref] 
  11. [11] Chebana, F. and T. Ouarda (2011). Multivariate quantiles in hydrological frequency analysis. Environmetrics 22(1), 63–78. [Crossref] 
  12. [12] Coles, S., J. Heffernan, and J. Tawn (1999). Dependence measures for extremes value analyses. Extremes 2(4), 339–365. [Crossref] Zbl0972.62030
  13. [13] Cont, R. (2009). La statistique face aux événements rares. Pour la science 385, 116–123. 
  14. [14] Deheuvels, P. (1979). La fonction de dépendance empirique et ses propriétés. Un test non paramétrique d’indépendance. Acad. Roy. Belg. Bull. Cl. Sci. 5(65), 274–292. Zbl0422.62037
  15. [15] Dobric, J. and F. Schmid (2005). Nonparametric estimation of the lower tail dependence λL in bivariate copulas. J. Appl. Statist. 32, 387–407. [Crossref] Zbl1121.62364
  16. [16] Draisma, G., H. Drees, A. Ferreira, and L. de Haan (2004). Bivariate tail estimation: dependence in asymptotic independence. Bernoulli 10(2), 251–280. [Crossref] Zbl1058.62043
  17. [17] Drees, H. and X. Huang (1998). Best attainable rates of convergence for estimators of the stable tail dependence function. J. Multivariate Anal. 64(1), 25–47. [Crossref] Zbl0953.62046
  18. [18] Dupuis, D. J. (2007). Using copulas in hydrology: benefits, cautions, and issues. J. Hydrol. Eng. 12(4), 381–393. [Crossref] 
  19. [19] Durante, F., J. Fernández-Sánchez, and R. Pappadà (2015). Copulas, diagonals, and tail dependence. Fuzzy Set. Syst. 264, 22–41. 
  20. [20] Einmahl, J. H. J., A. Krajina, and J. Segers (2008). A method of moments estimator of tail dependence. Bernoulli 14, 1003– 1026. [Crossref] Zbl1155.62017
  21. [21] Embrechts, P., A. McNeil, and D. Straumann (1999). Correlation: Pitfalls and alternatives. Risk Magazine May, 69–71. 
  22. [22] Embrechts, P., A. Mcneil, and D. Straumann (2002). Correlation and dependence in riskmanagement: Properties and pitfalls. In Risk Management: Value at Risk and Beyond, M.A.H. Dempster ed., Cambridge University Press, 176-223. 
  23. [23] Fischer, M. and M. Dörflinger (2006). A note on a non-parametric tail dependence estimator. Discussion Papers 76/2006, Friedrich-Alexander-University Erlangen-Nuremberg. 
  24. [24] Frahm, G., M. Junker, and R. Schmidt (2005). Estimating the tail-dependence coefficient: properties and pitfalls. Insurance Math. Econ. 37(1), 80–100. Zbl1101.62012
  25. [25] Galambos, J. (1978). The Asymptotic Theory of Extreme Order Statistics. Wiley, New York. Zbl0381.62039
  26. [26] Geffroy, J., 1958: Contribution à la théorie des valeurs extrêmes. Public. Inst. Stat. Paris 7 7, 37–121. 
  27. [27] Geffroy, J. (1959). Contribution à la théorie des valeurs extrêmes. Public. Inst. Stat. Paris 7 8, 123–184. Zbl0092.34901
  28. [28] Genest, C. and F. Chebana (2015). Copula modeling in hydrological frequency analysis. To appear in Chow’s Handbook of Applied Hydrology, Second Edition, McGraw-Hill. 
  29. [29] Genest, C. and A.-C. Favre (2007). Everything you always wanted to know about copula modeling but were afraid to ask. J. Hydrol. Eng. 12(4), 347–368. [Crossref] 
  30. [30] Genest, C., I. Kojadinovic, J. Nešlehová, and J. Yan (2010). A goodness-of-fit test for bivariate extreme-value copulas. Bernoulli 17(1), 253–275. [Crossref] Zbl1284.62331
  31. [31] Genest, C. and J. Segers (2009). Rank-based inference for bivariate extreme-value copulas. Ann. Stat. 37, 2990–3022. [Crossref] Zbl1173.62013
  32. [32] Ghosh, S. (2010). Modelling bivariate rainfall distribution and generating bivariate correlated rainfall data in neighbouring meteorological subdivisions using copula. Hydrol. Process. 24(24), 3558–3567. 
  33. [33] Ghoudi, K., A. Khoudraji, and L.-P. Rivest (1998). Propriétés statistiques des copules de valeurs extrêmes bidimensionnelles. Canad. J.Stat. 26(1), 187–197. 
  34. [34] Heffernan, J. (2000). A directory of coefficients of tail dependence. Extremes 3(3), 279–290. [Crossref] Zbl0979.62040
  35. [35] Hill, B. (1975). A simple general approach to inference about the tail of a distribution. Ann. Stat. 3(5), 1163–1174. [Crossref] Zbl0323.62033
  36. [36] Huang, X. (1992). Statistics of Bivariate Extreme Values. Ph.D. thesis, Erasmus University Rotterdam. 
  37. [37] Joe, H. (1997). Multivariate Models and Dependence Concepts. Chapman and Hall, London. Zbl0990.62517
  38. [38] Joe, H., R. L. Smith, and I. Weissman (1992). Bivariate threshold methods for extremes. J. R. Stat. Soc. Ser. B 54(1), 171–183. Zbl0775.62083
  39. [39] Juri, A. and M. Wuthrich (2002). Copula convergence theorems for tail events. Insurance Math. Econ. 30(3), 405–420. Zbl1039.62043
  40. [40] Kendall, M. G. (1938). A new measure of rank correlation. Biometrika 30(1/2), 81–93. [Crossref] Zbl0019.13001
  41. [41] Kojadinovic, I. and J. Yan (2010). Nonparametric rank-based tests of bivariate extreme-value dependence. J. Multivariate Anal. 101, 2234–2249. [Crossref] Zbl1201.62056
  42. [42] Kratz, M. and S. Resnick (1996). The QQ-estimator and heavy tails. Stochastic Models 12(4), 699–724. Zbl0887.62025
  43. [43] Ledford, A. W. and J. A. Tawn (1996). Statistics for near independence in multivariate extreme values. Biometrika 83(1), 169–187. [Crossref] Zbl0865.62040
  44. [44] Ledford, A. W. and J. A. Tawn (1997). Modelling dependence within joint tail regions. J. R. Stat. Soc. Ser. B 59(2), 475–499. [Crossref] Zbl0886.62063
  45. [45] Ledford, A. W. and J. A. Tawn (1998). Concomitants of extremes. Adv. Appl. Probab. 30(1), 197–215. [Crossref] Zbl0905.60034
  46. [46] Lee, T., R. Modarres, and T. Ouarda (2013). Data based analysis of bivariate copula tail dependence for drought duration and severity. Hydrol. Process. 27(10), 1454–1463. [Crossref] 
  47. [47] Lekina, A., (2010). Estimation Non-paramétrique des Quantiles Extrêmes Conditionnels. Ph.D. thesis, Université de Grenoble. 
  48. [48] Lekina, A., F. Chebana, and T. Ouarda (2014). Weighted estimate of extreme quantile: an application to the estimation of high flood return periods. Stoch. Env. Res. Risk. A. 28(2), 147–165. [Crossref] 
  49. [49] Longin, F. M. and B. Solnik (2001). Extreme correlations of international equity markets during extremely volatile periods. J. Finance 56, 649–676. [Crossref] 
  50. [50] Malevergne, Y. and D. Sornette (2004). Investigating extreme dependences. Review Financ. Studies, in press. Zbl1093.62098
  51. [51] Nelsen, R. B. (2006). An Introduction to Copulas, Second edition, Springer-Verlag New York. Zbl1152.62030
  52. [52] Ouarda, T. B. M. J., M. Haché, P. Bruneau, and B. Bobée (2000). Regional flood peak and volume estimation in northern canadian basin. J. Cold Reg. Eng. 14(4), 176–191. 
  53. [53] Patton, A. (2006). Estimation of multivariate models for time series of possibly different lengths. J. Appl. Econom. 21(2), 147–173. [Crossref] 
  54. [54] Peng, L. (1999). Estimation of the coefficient of tail dependence in bivariate extremes. Stat. Probabil. Lett. 43(4), 399–409. [Crossref] Zbl0958.62049
  55. [55] Pickands, J. (1975). Statistical inference using extreme order statistics. Ann. Stat. 3(1), 119–131. [Crossref] Zbl0312.62038
  56. [56] Pickands, J. (1981). Multivariate extreme value distributions. Bull. Inst. Internat. Statist. 49(2), 859–878. Zbl0518.62045
  57. [57] Poon, S.-H., M. Rockinger, and J. Tawn (2004). Extreme value dependence in financial markets: diagnostics, models, and financial implications. Review Financ. Studies 17(2), 581–610. [Crossref] 
  58. [58] Poulin, A., D. Huard, A.-C. Favre, and S. Pugin (2007). Importance of tail dependence in bivariate frequency analysis. J. Hydrol. Eng. 12(4), 394–403. [Crossref] 
  59. [59] Raynal-Villasenor, J. and J. Salas (1987). Multivariate extreme value distributions in hydrological analyses. In Water for the Future: Hydrology in Perspective, Rodda, J. C. and N.C. Matalas eds., 111–119. 
  60. [60] Resnick, S. (1987). Extreme Values, Regular Variation, and Point Process. Springer Verlag, New-York. Zbl0633.60001
  61. [61] Rohatgi, V. (1976). An Introduction to Probability Theory and Mathematical Statistics. John Wiley & Sons, New York. Zbl0354.62001
  62. [62] Salvadori, G. and C. De Michele (2011). Estimating strategies for multiparameter multivariate extreme value copulas. Hydrol. Earth Syst. Sc. 15(1), 141–150. [Crossref] 
  63. [63] Salvadori, G., C. De Michele, N. T. Kottegoda, and R. Rosso (2007). Extremes in Nature: an Approach Using Copulas. Springer, Dordrecht. 
  64. [64] Schmidt, R. (2002). Tail dependence for elliptically contoured distributions. Math. Methods Oper. Res. 55, 301–327. [Crossref] Zbl1015.62052
  65. [65] Schmidt, R. and U. Stadtmuller (2006). Non-parametric estimation of tail dependence. Scand. J. Statist. 33, 307–335. [Crossref] Zbl1124.62016
  66. [66] Schultze, J. and J. Steinebach (1996). On least squares estimates of an exponential tail coefficient. Stat. Decisions 14(3), 353–372. Zbl0893.62023
  67. [67] Segers, J. (2007). Non-parametric inference for bivariate extreme-value copulas. In Topics in Extreme Values, M. Ahsanullah and S. N. U. A. Kirmani eds., Nova Science Publishers Inc, New York, 181–203. 
  68. [68] Serinaldi, F. (2008). Analysis of inter-gauge dependence by Kendall’s fiK, upper tail dependence coefficient, and 2-copulas with application to rainfall fields. Stoch. Env. Res. Risk. A. 22(6), 671–688. 
  69. [69] Shiau, J. T. (2003). Return period of bivariate distributed extreme hydrological events. Stoch. Env. Res. Risk. A. 17, 42–57. [Crossref] Zbl1025.62048
  70. [70] Shiau, J.-T., R. Modarres, and S. Nadarajah (2012). Assessing multi-site drought connections in Iran using empirical copula. Environ. Model. Assess. 17(5), 469–482. [Crossref] 
  71. [71] Sklar, A. (1959). Fonstions de répartition à n dimensions et leurs marges. Publ. Inst. Statist. Univ. Paris 8, 229–231. Zbl0100.14202
  72. [72] Venter, G. G., 2002: Tails of copulas. Proc. Casualty Actuarial Soc. 89, 68–113. 
  73. [73] Wasserman, L. (2003). All of Statistics: a Concise Course in Statistical Inference. Springer, New York. Zbl1053.62005
  74. [74] Yue, S., T. Ouarda, B. Bobée, P. Legendre, and P. Bruneau (1999). The Gumbel mixed model for flood frequency analysis. J. Hydrol. 226(1-2), 88–100. [Crossref] 
  75. [75] Zhang, L. and V. P. Singh (2006). Bivariate flood frequency analysis using the copula method. J. Hydrol. Eng. 11(2), 150–164. [Crossref] 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.