Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models

Christian Genest; Bruno Rémillard

Annales de l'I.H.P. Probabilités et statistiques (2008)

  • Volume: 44, Issue: 6, page 1096-1127
  • ISSN: 0246-0203

Abstract

top
In testing that a given distribution Pbelongs to a parameterized family 𝒫 , one is often led to compare a nonparametric estimateAn of some functional A of P with an element Aθn corresponding to an estimate θn of θ. In many cases, the asymptotic distribution of goodness-of-fit statistics derived from the process n1/2(An−Aθn) depends on the unknown distribution P. It is shown here that if the sequences An and θn of estimators are regular in some sense, a parametric bootstrap approach yields valid approximations for the P-values of the tests. In other words if An* and θn* are analogs of An and θn computed from a sample from Pθn, the empirical processes n1/2(An−Aθn) and n1/2(An*−Aθn*) then converge jointly in distribution to independent copies of the same limit. This result is used to establish the validity of the parametric bootstrap method when testing the goodness-of-fit of families of multivariate distributions and copulas. Two types of tests are considered: certain procedures compare the empirical version of a distribution function or copula and its parametric estimation under the null hypothesis; others measure the distance between a parametric and a nonparametric estimation of the distribution associated with the classical probability integral transform. The validity of a two-level bootstrap is also proved in cases where the parametric estimate cannot be computed easily. The methodology is illustrated using a new goodness-of-fit test statistic for copulas based on a Cramér–von Mises functional of the empirical copula process.

How to cite

top

Genest, Christian, and Rémillard, Bruno. "Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models." Annales de l'I.H.P. Probabilités et statistiques 44.6 (2008): 1096-1127. <http://eudml.org/doc/78005>.

@article{Genest2008,
abstract = {In testing that a given distribution Pbelongs to a parameterized family $\mathcal \{P\}$, one is often led to compare a nonparametric estimateAn of some functional A of P with an element Aθn corresponding to an estimate θn of θ. In many cases, the asymptotic distribution of goodness-of-fit statistics derived from the process n1/2(An−Aθn) depends on the unknown distribution P. It is shown here that if the sequences An and θn of estimators are regular in some sense, a parametric bootstrap approach yields valid approximations for the P-values of the tests. In other words if An* and θn* are analogs of An and θn computed from a sample from Pθn, the empirical processes n1/2(An−Aθn) and n1/2(An*−Aθn*) then converge jointly in distribution to independent copies of the same limit. This result is used to establish the validity of the parametric bootstrap method when testing the goodness-of-fit of families of multivariate distributions and copulas. Two types of tests are considered: certain procedures compare the empirical version of a distribution function or copula and its parametric estimation under the null hypothesis; others measure the distance between a parametric and a nonparametric estimation of the distribution associated with the classical probability integral transform. The validity of a two-level bootstrap is also proved in cases where the parametric estimate cannot be computed easily. The methodology is illustrated using a new goodness-of-fit test statistic for copulas based on a Cramér–von Mises functional of the empirical copula process.},
author = {Genest, Christian, Rémillard, Bruno},
journal = {Annales de l'I.H.P. Probabilités et statistiques},
keywords = {copula; goodness-of-fit test; Monte Carlo simulation; parametric bootstrap; P-values; semiparametric estimation; -values},
language = {eng},
number = {6},
pages = {1096-1127},
publisher = {Gauthier-Villars},
title = {Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models},
url = {http://eudml.org/doc/78005},
volume = {44},
year = {2008},
}

TY - JOUR
AU - Genest, Christian
AU - Rémillard, Bruno
TI - Validity of the parametric bootstrap for goodness-of-fit testing in semiparametric models
JO - Annales de l'I.H.P. Probabilités et statistiques
PY - 2008
PB - Gauthier-Villars
VL - 44
IS - 6
SP - 1096
EP - 1127
AB - In testing that a given distribution Pbelongs to a parameterized family $\mathcal {P}$, one is often led to compare a nonparametric estimateAn of some functional A of P with an element Aθn corresponding to an estimate θn of θ. In many cases, the asymptotic distribution of goodness-of-fit statistics derived from the process n1/2(An−Aθn) depends on the unknown distribution P. It is shown here that if the sequences An and θn of estimators are regular in some sense, a parametric bootstrap approach yields valid approximations for the P-values of the tests. In other words if An* and θn* are analogs of An and θn computed from a sample from Pθn, the empirical processes n1/2(An−Aθn) and n1/2(An*−Aθn*) then converge jointly in distribution to independent copies of the same limit. This result is used to establish the validity of the parametric bootstrap method when testing the goodness-of-fit of families of multivariate distributions and copulas. Two types of tests are considered: certain procedures compare the empirical version of a distribution function or copula and its parametric estimation under the null hypothesis; others measure the distance between a parametric and a nonparametric estimation of the distribution associated with the classical probability integral transform. The validity of a two-level bootstrap is also proved in cases where the parametric estimate cannot be computed easily. The methodology is illustrated using a new goodness-of-fit test statistic for copulas based on a Cramér–von Mises functional of the empirical copula process.
LA - eng
KW - copula; goodness-of-fit test; Monte Carlo simulation; parametric bootstrap; P-values; semiparametric estimation; -values
UR - http://eudml.org/doc/78005
ER -

References

top
  1. [1] P. Barbe, C. Genest, K. Ghoudi and B. Rémillard. On Kendall’s process. J. Multivariate Anal. 58 (1996) 197–229. Zbl0862.60020MR1405589
  2. [2] R. Beran. Minimum distance procedures. In Nonparametric Methods 741–754. Handbook of Statistics 4. North-Holland, Amsterdam, 1984. Zbl0597.62032MR831734
  3. [3] R. Beran and P. W. Millar. A stochastic minimum distance test for multivariate parametric models. Ann. Statist. 17 (1989) 125–140. Zbl0684.62041MR981440
  4. [4] P. J. Bickel and J.-J. Ren. The bootstrap in hypothesis testing. In State of the Art in Probability and Statistics (Leiden, 1999) 91–112. IMS Lecture Notes Monogr. Ser. 36. Inst. Math. Statist., Beachwood, OH, 2001. MR1836556
  5. [5] P. J. Bickel and M. J. Wichura. Convergence criteria for multiparameter stochastic processes and some applications. Ann. Math. Statist. 42 (1971) 1656–1670. Zbl0265.60011MR383482
  6. [6] W. Breymann, A. Dias and P. Embrechts. Dependence structures for multivariate high-frequency data in finance. In Selected Proceedings from Quantitative Methods in Finance, 2002 (Cairns/Sydney) 3 1–14, 2003. MR1972372
  7. [7] S. Demarta and A. J. McNeil. The t copula and related copulas. Internat. Statist. Rev. 73 (2005) 111–129. Zbl1104.62060
  8. [8] J. Dobrić and F. Schmid. A goodness of fit test for copulas based on Rosenblatt’s transformation. Comput. Statist. Data Anal. 51 (2007) 4633–4642. Zbl1162.62343MR2364470
  9. [9] J. Durbin. Weak convergence of the sample distribution function when parameters are estimated. Ann. Statist. 1 (1973) 279–290. Zbl0256.62021MR359131
  10. [10] J.-D. Fermanian. Goodness-of-fit tests for copulas. J. Multivariate Anal. 95 (2005) 119–152. Zbl1095.62052MR2164126
  11. [11] J.-D. Fermanian, D. Radulović and M. H. Wegkamp. Weak convergence of empirical copula processes. Bernoulli 10 (2004) 847–860. Zbl1068.62059MR2093613
  12. [12] P. Gänßler and W. Stute. Seminar on Empirical Processes. Birkhäuser Verlag, Basel, 1987. Zbl0637.62047MR902803
  13. [13] C. Genest, K. Ghoudi and L.-P. Rivest. A semiparametric estimation procedure of dependence parameters in multivariate families of distributions. Biometrika 82 (1995) 543–552. Zbl0831.62030MR1366280
  14. [14] C. Genest, J.-F. Quessy and B. Rémillard. Tests of serial independence based on Kendall’s process. Canad. J. Statist. 30 (2002) 441–461. Zbl1016.62051MR1944373
  15. [15] C. Genest, J.-F. Quessy and B. Rémillard. Goodness-of-fit procedures for copula models based on the probability integral transformation. Scand. J. Statist. 33 (2006) 337–366. Zbl1124.62028MR2279646
  16. [16] C. Genest, B. Rémillard and D. Beaudoin. Goodness-of-fit tests for copulas: A review and a power study. Insurance Math. Econom. 43 (2008). In press. Zbl1161.91416MR2517885
  17. [17] C. Genest and L.-P. Rivest. Statistical inference procedures for bivariate Archimedean copulas. J. Amer. Statist. Assoc. 88 (1993) 1034–1043. Zbl0785.62032MR1242947
  18. [18] K. Ghoudi and B. Rémillard. Empirical processes based on pseudo-observations. In Asymptotic Methods in Probability and Statistics (Ottawa, ON, 1997) 171–197. North-Holland, Amsterdam, 1998. Zbl0959.62044MR1661480
  19. [19] K. Ghoudi and B. Rémillard. Empirical processes based on pseudo-observations. II. The multivariate case. In Asymptotic Methods in Stochastics 381–406. Fields Inst. Commun. 44. Amer. Math. Soc., Providence, RI, 2004. Zbl1079.60024MR2106867
  20. [20] N. Henze. Empirical-distribution-function goodness-of-fit tests for discrete models. Canad. J. Statist. 24 (1996) 81–93. Zbl0846.62037MR1394742
  21. [21] M. N. Jouini and R. T. Clemen. Copula models for aggregating expert opinions. Oper. Res. 44 (1996) 444–457. Zbl0864.90067
  22. [22] C. A. J. Klaassen and J. A. Wellner. Efficient estimation in the bivariate normal copula model: Normal margins are least favourable. Bernoulli 3 (1997) 55–77. Zbl0877.62055MR1466545
  23. [23] Y. Malevergne and D. Sornette. Testing the Gaussian copula hypothesis for financial assets dependences. Quant. Finance 3 (2003) 231–250. MR1999654
  24. [24] D. Pollard. The minimum distance method of testing. Metrika 27 (1980) 43–70. Zbl0425.62029MR563412
  25. [25] J. H. Shih and T. A. Louis. Inferences on the association parameter in copula models for bivariate survival data. Biometrics 51 (1995) 1384–1399. Zbl0869.62083MR1381050
  26. [26] W. Stute, W. González-Manteiga and M. Presedo-Quindimil. Bootstrap based goodness-of-fit tests. Metrika 40 (1993) 243–256. Zbl0770.62016MR1235086
  27. [27] H. Tsukahara. Semiparametric estimation in copula models. Canad. J. Statist. 33 (2005) 357–375. Zbl1077.62022MR2193980
  28. [28] A. W. van der Vaart and J. A. Wellner. Weak Convergence and Empirical Processes. Springer, New York, 1996. Zbl0862.60002MR1385671
  29. [29] W. Wang and M. T. Wells. Model selection and semiparametric inference for bivariate failure-time data (with discussion). J. Amer. Statist. Assoc. 95 (2000) 62–76. Zbl0996.62091MR1803141

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.