Data-driven score tests for homoscedastic linear regression model : asymptotic results

• Autorzy: Inglot T. , Ledwina T.
• Języki publikacji: EN

Abstrakty

EN

We describe and investigate new tests for testing the validity of a semiparametric random-design linear regression model. The tests were introduced in Inglot and Ledwina (2006a, b). We repeat here basic steps of the constructions. The resulting statistics are closely linked to some norms of the appropriate efficient score vector and related quantities. A useful way of deriving the efficient score vector is proposed and discussed. We introduce also a large class of estimators of the efficient score vector and prove that under the null model our constructions are asymptotically distribution free. The proof adopts and exploits some ideas and results developed in the area of semiparametric estimation. We give also the limiting distribution of the test statistics under the null hypothesis. The simulation results contained in Inglot and Ledwina (2006a, b) show the very good performance of the proposed tests.

Słowa kluczowe

EN

efficient score; hypothesis testing; data-driven test; linear regression; selection rule; semiparametric inference; smoothing methods

• Wydawca: Wrocław University of Science and Technology
• Czasopismo: Probability and Mathematical Statistics
• Rocznik: 2006
• Tom: Vol. 26, Fasc. 1
• Strony: 41--61
• Opis fizyczny: Bibliogr. 53 poz.

Twórcy

autor

Inglot T.

• Institute of Mathematics and Informatics, Wrocław University of Technology Wybrzeże Wyspiańskiego 27, 50-370 Wrocław, Poland
• Polish Academy of Sciences

autor

Ledwina T.

• Institute of Mathematics, Polish Academy of Sciences, ul. Kopernika 18, 51-617 Wrocław, Poland

Bibliografia

• [1] M. Aerts, G. Claeskens and J. D. Hart, Testing lack of fit in multiple regression, Biometrika 87 (2000), pp. 405-24.
• [2] A. Azzalini and A. Bowman, On use of nonparametric regression for checking linear relationships, J. Roy. Statist. Soc. В 55 (1993), pp. 549-557.
• [3] Y. Baraud, S. Huet and B. Laurent, Adaptive tests of linear hypotheses by model selection, Ann. Statist. 31 (2003), pp. 225-251.
• [4] P. J. Bickel, On adaptive estimation, Ann. Statist 10 (1982), pp. 647-671.
• [5] P. J. Вiскel, С. А. К1aassen, Y. Rittо V and J. A. Wellner, Efficient and Adaptive Estimation for Semiparametric Models, John Hopkins Univ. Press, Baltimore 1993.
• [6] P. J. Bickel, Y. Rito V and T. Stocker, Testing and the method of sieves, Technical Report, University of California, Berkeley, 1998.
• [7] S, Choi, On asymptotically optimal tests, Ph.D. Thesis, Dept. Statistics, University of Rochester, Rochester, 1989.
• [8] S. Choi, W. J. Hall and A. Schick, Asymptotically uniformly most powerful tests in parametric and semiparametric models, Ann. Statist. 24 (1996), pp. 841-861.
• [9] D. Cox, E. Кoh, G. Wahba and B. Yandell, Testing the (parametric) null model hypothesis in (semiparametric) partial and generalized spline models, Ann. Statist. 16 (1988), pp. 113-119.
• [10] H. Dette, On a nonparametric test for linear relationships, Statist. Probab. Lett. 46 (2000), pp. 307-316.
• [11] J. Diebolt and J. Zuber, On the half-sample method for goodness-of-fit in regression, Technical Report 00.1, Department of Mathematics and Statistics, La Trobe University, 2000.
• [12] R. L. Eubank, J. D. Hart and V. N. LaRiccia, Testing goodness of fit via nonparametric function estimation techniques, Comm. Statist. Theory Methods 22 (1993), pp. 3327-3354.
• [13] J. Fan, Test of significance based on wavelet thresholding and Neyman’s truncation, J. Amer. Statist, Assoc. 91 (1996), pp. 674-688.
• [14] J. Fan and L.-S. Huang, Goodness-of-fit tests for parametric regression models, J. Amer. Statist. Assoc. 96 (2001), pp. 640-652.
• [15] E. Guerre and P. Lavergne, Data-driven rate-optimal specification testing in regression models, Ann. Statist. 33 (2005), pp. 840-870.
• [16] J. Hájek and Z. Sidák, Theory of Rank Tests, Academia, Prague 1967.
• [17] W. Härdle and E. Mammen, Comparing nonparametric versus parametric regression fits, Ann. Statist. 21 (1993), pp. 1926-1947.
• [18] J. D. Hart, Nonparametric Smoothing and Lack-of-fit Tests, Springer, New York 1997.
• [19] J. L. Horowitz and V. G. Spokoiny, An adaptive, rate optimal test of a parametric mean-regression model against a nonparametric alternative, Econometrica 69 (2001), pp. 599-631.
• [20] T. Inglot, W. C. M. Kallenberg and T. Ledwina, Data driven smooth tests for composite hypotheses, Ann. Statist. 25 (1997), pp. 1222-1250.
• [21] T. Inglot and T. Ledwina, Asymptotic optimality of data driven Neyman’s tests for uniformity, Ann. Statist. 24 (1996), pp. 1982-2019.
• [22] T. Inglot and T. Ledwina, Intermediate approach to comparison of some goodness-of-fit tests, Ann. Inst. Statist. Math. 53 (2001a), pp. 810-834.
• [23] T. Inglot and T. Ledwina, Asymptotic optimality of data driven smooth tests for location-scale family, Sankhya, Ser. A 63 (2001b), pp. 41-71.
• [24] T. Inglot and T. Ledwina, Semiparametric regression: Hadamard differentiability and efficient score functions for some testing problems, Preprint 012, Institute of Mathematics, Wrocław University of Technology, 2003.
• [25] T. Inglot and T. Ledwina, Data driven score tests for homoscedastic linear regression model, Preprint 665, Institute of Mathematics of the Polish Academy of Sciences, 2006a.
• [26] T. Inglot and T. Ledwina, Data driven score tests for homoscedastic linear regression model: the construction and simulations, in: Prague Stochastics 2006, Proceedings, M. Husková and M. Janžura (Eds.), matfyzpress, Prague 2006b, pp. 124-137.
• [27] T. Inglot and T. Ledwina, Intermediate efficiency of some max-type statistics, J. Statist. Plann. Inference 136 (2006c), pp. 2918-2935.
• [28] T. Inglot and T. Ledwina, Towards data driven selection of a penalty function for data driven Neyman tests, Linear Algebra Appl. 417 (2006d), pp. 124-133.
• [29] T. Inglot and T. Ledwina, Asymptotic optimality of new adaptive test in regression model, Ann. Inst. H. Poincare (2006e).
• [30] H. S. Javitz, Generalized smooth tests of goodness of fit, independence and equality of distributions, Ph. D. Thesis, Univ. California, Berkeley, 1975.
• [31] W. C. M. Kallenberg and T. Ledwina, On data driven Neyman s tests, Probab. Math. Statist. 15 (1995), pp. 409-426.
• [32] W. C. M. Kallenberg and T. Ledwina, Data-driven smooth tests when the hypothesis is composite, J. Amer. Statist. Assoc. 92 (1997), pp. 1094-1104.
• [33] E. V. Khmaladze and H. L. Koul, Martingale transforms goodness-of-fit tests in regression models, Ann. Statist. 32 (2004), pp. 995-1034.
• [34] R. Koenker and Z. Xiao, Inference on the quantile regression process, Econometrica 70 (2002), pp. 1583-1612.
• [35] H. L. Koul and A. Schick, Adaptive estimation in a random coefficient autoregressive model, Ann. Statist. 24 (1996), pp. 1025-1052.
• [36] A. S. Kozek, A nonparametric test of fit of a parametric model, J. Multivariate Anal. 37 (1991), pp. 66-75.
• [37]J.-P. Kreiss, On adaptive estimation in stationary ARMA processes, Ann. Statist. 15 (1987), pp. 112-133. _
• [38] L. Le Cam, On the asymptotic theory of estimation and testing hypotheses, Proc. Third Berkeley Symp. Math. Statist. Probab. 1, Univ. California Press, Berkeley 1956, pp. 129-156.
• [39] L. Le Cam and E. L. Lehmann, J. Neyman-On the occasion of his 80th birthday, Ann. Statist. 2 (1974), pp. vii-xiii.
• [40] T. Ledwina, Data driven version of Neyman’s smooth test of fit, J. Amer. Statist. Assoc. 89 (1994), pp. 1000-1005.
• [41] J. Neyman, “Smooth test” for goodness of fit, Skand. Aktuarietidskr. 20 (1937), pp. 149-199.
• [42] J. Neyman, Sur une familie de tests asymptotiques des hypotheses statistiques composées, Trabajos de Estadistica 5 (1954), pp. 161-168.
• [43] J. Neyman, Optimal asymptotic tests of composite statistical hypotheses, in: Probability and Statistics: The Harald Cramér Volume, U. Grenander (Ed.), Wiley, New York 1959, pp. 213-234.
• [44] J. C. W. Rayner and D. J. Best, Smooth Tests of Goodness of Fit, Oxford Univ. Press, New York 1989.
• [45] A. Schick, On asymptotically efficient estimation in semiparametric models, Ann. Statist 14 (1986), pp. 1139-1151.
• [46] A. Schick, Efficient estimates in linear and nonlinear regression with heteroscedastic error, J. Statist. Plann. Inference 58 (1997), pp. 371-387.
• [47] A. Schick, On asymptotic differentiability of averages, Statist. Probab. Lett. 51 (2001), pp. 15-23.
• [48] J. S. Simonoff and C. L. Tsai, Semiparametric and additive model selection using an improved Akaike Information Criterion, J. Comput. Graph. Statist. 8 (1999), pp. 22-40.
• [49] W. Stute, Nonparametric model checks for regression, Ann. Statist. 25 (1997), pp. 613-641.
• [50] W. Stute, W. G. Manteiga and M. P. Quindimil, Bootstrap approximation in model checks in regression, J. Amer. Statist. Assoc. 93 (1998a), pp. 141-149.
• [51] W. Stute, S. Thies and L.-X. Zhu, Model checks for regression: an innovation process approach, Ann. Statist. 26 (1998b), pp. 1916-1934.
• [52] D. Thomas and D. Pierce, Neyman’s smooth goodness-of-fit test when the hypothesis is composite, J. Amer. Statist. Assoc. 74 (1979), pp. 441-445.
• [53] A. W. van der Vaart, On differentiable functionals, Ann. Statist. 19 (1991), pp. 178-204.