zbMATH — the first resource for mathematics

Generalized likelihood ratio statistics and Wilks phenomenon. (English) Zbl 1029.62042
Summary: Likelihood ratio theory has had tremendous success in parametric inference, due to the fundamental theory of Wilks. Yet, there is no general applicable approach for nonparametric inferences based on function estimation. Maximum likelihood ratio test statistics in general may not exist in nonparametric function estimation settings. Even if they exist, they are hard to find and can not be optimal as shown in this paper.
We introduce generalized likelihood statistics to overcome the drawbacks of nonparametric maximum likelihood ratio statistics. A new Wilks phenomenon is unveiled. We demonstrate that a class of the generalized likelihood statistics based on some appropriate nonparametric estimators are asymptotically distribution free and follow \(\chi^2\)-distributions under null hypotheses for a number of useful hypotheses and a variety of useful models including Gaussian white noise models, nonparametric regression models, varying coefficient models and generalized varying coefficient models.
We further demonstrate that generalized likelihood ratio statistics are asymptotically optimal in the sense that they achieve optimal rates of convergence. They can even be adaptively optimal by using a simple choice of adaptive smoothing parameters. Our work indicates that the generalized likelihood ratio statistics are indeed general and powerful for nonparametric testing problems based on function estimation.

62G10 Nonparametric hypothesis testing
62G07 Density estimation
62G20 Asymptotic properties of nonparametric inference
62J12 Generalized linear models (logistic models)
Full Text: DOI
[1] Aerts, M., Claeskens, G. and Hart, J. D. (1999). Testing the fit of a parametric function. J. Amer. Statist. Assoc. 94 869-879. JSTOR: · Zbl 0996.62044 · doi:10.2307/2670002 · links.jstor.org
[2] Azzalini, A. and Bowman, A. N. (1993). On the use of nonparametric regression for checking linear relationships. J. Roy. Statist. Soc. Ser. B 55 549-557. JSTOR: · Zbl 0800.62222 · links.jstor.org
[3] Azzalini, A., Bowman, A. N. and Härdle, W. (1989). On the use of nonparametric regression for model checking. Biometrika 76 1-11. JSTOR: · Zbl 0663.62096 · doi:10.1093/biomet/76.1.1 · links.jstor.org
[4] Bickel, P. J. and Ritov, Y. (1992). Testing for goodness of fit: a new approach. In Nonparametric Statistics and Related Topics (A. K. Md. E. Saleh, ed.) 51-57. North-Holland, New York.
[5] Bickel, P. J. and Rosenblatt, M. (1973). On some global measures of the deviation of density function estimates. Ann. Statist. 1 1071-1095. · Zbl 0275.62033 · doi:10.1214/aos/1176342558
[6] Brown, L. D. and Low, M. G. (1996). Asymptotic equivalence of nonparametric regression and white noise. Ann. Statist. 24 2384-2398. · Zbl 0867.62022 · doi:10.1214/aos/1032181159
[7] Cai, Z., Fan, J. and Li, R. (2000). Efficient estimation and inferences for varying-coefficient models. J. Amer. Statist. Assoc. To appear. JSTOR: · Zbl 0999.62052 · doi:10.2307/2669472 · links.jstor.org
[8] Carroll, R. J., Fan, J., Gijbels, I. and Wand, M. P. (1997). Generalized partially linear singleindex models. J. Amer. Statist. Assoc. 92 477-489. JSTOR: · Zbl 0890.62053 · doi:10.2307/2965697 · links.jstor.org
[9] Chen, J. H. and Qin, J. (1993). Empirical likelihood estimation for finite populations and the effective usage of auxiliary information. Biometrika 80 107-116. JSTOR: · Zbl 0769.62006 · doi:10.1093/biomet/80.1.107 · links.jstor.org
[10] Cleveland, W. S. and Devlin, S. J. (1988). Locally-weighted regression: an approach to regression analysis by local fitting. J. Amer. Statist. Assoc. 83 597-610. · Zbl 1248.62054
[11] Cleveland, W. S., Grosse, E. and Shyu, W. M. (1992). Local regression models. In Statistical Models in S (J. M. Chambers and T. J. Hastie, eds.) 309-376. Wadsworth and Brooks Cole, Pacific Grove, CA.
[12] de Jong, P. (1987). A central limit theorem for generalized quadratic forms. Probab. Theory Related Fields 75 261-277. · Zbl 0596.60022 · doi:10.1007/BF00354037
[13] Eubank, R. L. and Hart, J. D. (1992). Testing goodness-of-fit in regression via order selection criteria. Ann. Statist. 20 1412-1425. · Zbl 0776.62045 · doi:10.1214/aos/1176348775
[14] Eubank, R. L. and LaRiccia, V. M. (1992). Asymptotic comparison of Cramér-von Mises and nonparametric function estimation techniques for testing goodness-of-fit. Ann. Statist. 20 2071-2086. · Zbl 0769.62033 · doi:10.1214/aos/1176348903
[15] Fan, J. (1993). Local linear regression smoothers and their minimax efficiency. Ann. Statist. 21 196-216. · Zbl 0773.62029 · doi:10.1214/aos/1176349022
[16] Fan, J. (1996). Test of significance based on wavelet thresholding and Neyman’s truncation. J. Amer. Statist. Assoc. 91 674-688. JSTOR: · Zbl 0869.62032 · doi:10.2307/2291663 · links.jstor.org
[17] Fan, J. and Gijbels, I. (1996). Local Polynomial Modeling and Its Applications. Chapman and Hall, London. · Zbl 0873.62037
[18] Fan, J. and Huang, L. (1998). Goodness-of-fit test for parametric regression models. Technical Report, Dept. Statistics, Univ. California, Los Angeles.
[19] Fan, J. and Zhang, J. (1999). Sieve empirical likelihood ratios for nonparametric functions. Unpublished manuscript.
[20] Hall, P. and Owen, A. B. (1993). Empirical likelihood confidence bands in density estimation. J. Comput. Graph. Statist. 2 273-289. JSTOR: · Zbl 04516296 · doi:10.2307/1390646 · links.jstor.org
[21] Härdle, W. and Mammen, E. (1993). Comparing nonparametric versus parametric regression fits. Ann. Statist. 21 1926-1947. · Zbl 0795.62036 · doi:10.1214/aos/1176349403
[22] Hart, J. D. (1997). Nonparametric Smoothing and Lack-of-Fit Tests. Springer, New York. · Zbl 0886.62043
[23] Hastie, T. J. and Tibshirani, R. J. (1990). Generalized Additive Models. Chapman and Hall, London. · Zbl 0747.62061
[24] Hastie, T. J. and Tibshirani, R. J. (1993). Varying-coefficient models (with discussion). J. Royal Statist. Soc. Ser. B 55 757-796. JSTOR: · Zbl 0796.62060 · links.jstor.org
[25] Huber, P. J. (1973). Robust regression: asymptotics, conjectures and Monte Carlo. Ann. Statist. 1 799-821. · Zbl 0289.62033 · doi:10.1214/aos/1176342503
[26] Inglot, T., Kallenberg, W. C. M. and Ledwina, T. (1994). Power approximations to and power comparison of smooth goodness-of-fit tests. Scand. J. Statist. 21 131-145. · Zbl 0799.62042
[27] Inglot, T. and Ledwina, T. (1996). Asymptotic optimality of data-driven Neyman’s tests for uniformity. Ann. Statist. 24 1982-2019. · Zbl 0905.62044 · doi:10.1214/aos/1069362306
[28] Ingster, Yu. I. (1993). Asymptotic minimax hypothesis testing for nonparametric alternatives I-III. Math. Methods Statist. 2 85-114; 3 171-189; 4 249-268. · Zbl 0798.62059
[29] Kallenberg, W. C. M. and Ledwina, T. (1997). Data-driven smooth tests when the hypothesis is composite. J. Amer. Statist. Assoc. 92 1094-1104. JSTOR: · Zbl 1067.62534 · doi:10.2307/2965574 · links.jstor.org
[30] Koroljuk, V. S. and Borovskich, Yu. V. (1994). Theory of UStatistics. Kluwer, Amsterdam.
[31] Kuchibhatla, M. and Hart, J. D. (1996). Smoothing-based lack-of-fit tests: variations on a theme. J. Nonparameter. Statist. 7 1-22. · Zbl 0877.62041 · doi:10.1080/10485259608832685
[32] Lepski, O. V. and Spokoiny, V. G. (1999). Minimax nonparametric hypothesis testing: the case of an inhomogeneous alternative. Bernoulli 5 333-358. · Zbl 0946.62050 · doi:10.2307/3318439
[33] Li, G., Hollander, M., McKeague, I. W. and Yang, J. (1996). Nonparametric likelihood ratio confidence bands for quantile functions from incomplete survival data. Ann. Statist. 24 628-640. · Zbl 0859.62047 · doi:10.1214/aos/1032894455
[34] Murphy, S. A. (1993). Testing for a time dependent coefficient in Cox’s regression model. Scand. J. Statist. 20 35-50. · Zbl 0782.62053
[35] Neyman, J. (1937). Smooth test for goodness of fit. Skand. Aktuar. J. 20 149-199. · Zbl 0018.03403
[36] Nussbaum, M. (1996). Asymptotic equivalence of density estimation and Gaussian white noise. Ann. Statist. 24 2399-2430. · Zbl 0867.62035 · doi:10.1214/aos/1032181160
[37] Owen, A. B. (1988). Empirical likelihood ratio confidence intervals for a single functional. Biometrika 75 237-249. JSTOR: · Zbl 0641.62032 · doi:10.1093/biomet/75.2.237 · links.jstor.org
[38] Owen, A. B. (1990). Empirical likelihood ratio confidence regions. Ann. Statist. 18 90-120. · Zbl 0712.62040 · doi:10.1214/aos/1176347494
[39] Portnoy, S. (1988). Asymptotic behavior of likelihood methods for exponential families when the number of parameters tends to infinity. Ann. Statist. 16 356-366. · Zbl 0637.62026 · doi:10.1214/aos/1176350710
[40] Randle, D. H. and Wolfe, D. A. (1979). Introduction to the Theory of Nonparametric Statistics. Wiley, New York. · Zbl 0529.62035
[41] Seber, G. A. F. (1977). Linear Regression Analysis. Wiley, New York. · Zbl 0354.62055
[42] Shen, X., Shi, J. and Wong, W. H. (1999). Random sieve likelihood and general regression models. J. Amer. Statist. Assoc. 94 835-846. JSTOR: · Zbl 0994.62032 · doi:10.2307/2669998 · links.jstor.org
[43] Severini, T. A. and Wong, W. H. (1992). Generalized profile likelihood and conditional parametric models. Ann. Statist. 20 1768-1802. · Zbl 0768.62015 · doi:10.1214/aos/1176348889
[44] Silverman, B. W. (1984). Spline smoothing: the equivalent variable kernel method. Ann. Statist. 12 898-916. · Zbl 0547.62024 · doi:10.1214/aos/1176346710
[45] Spokoiny, V. G. (1996). Adaptive hypothesis testing using wavelets. Ann. Statist. 24 2477-2498. · Zbl 0898.62056 · doi:10.1214/aos/1032181163
[46] Wilks, S. S. (1938). The large-sample distribution of the likelihood ratio for testing composite hypotheses. Ann. Math. Statist. 9 60-62. · Zbl 0018.32003 · doi:10.1214/aoms/1177732360
[47] Zhang, J. and Gijbels, I. (1999). Sieve empirical likelihood and extensions of generalized least squares. Discussion paper, Institute of Statistics, Univ. catholique de Louvain.
[48] AZ, Eindhoven The Netherlands E-mail: jzhang@euridice.tue.nl
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. It attempts to reflect the references listed in the original paper as accurately as possible without claiming the completeness or perfect precision of the matching.