Likelihood ratio processes under nonstandard settings

Y. Goto   ; T. Kaneko   ; S. Kojima   ; M. Taniguchi

Teoriâ veroâtnostej i ee primeneniâ, Tome 67 (2022) no. 2, pp. 309-326

Résumé

This paper establishes the LAN property for the curved normal families and the simultaneous equation systems. In addition, we show that one-way random ANOVA models fail to have the LAN property. We consider the two cases when the variance of random effect lies in the interior and boundary of parameter space. In the former case, the log-likelihood ratio converges to $0$. In the latter case, the log-likelihood ratio has atypical limit distributions, which depend on the contiguity orders. The contiguity orders corresponding to the variances of random effects and disturbances can be equal to or greater than one, respectively, and that corresponding to the grand mean can be equal to or greater than one half. Consequently, we cannot use the ordinary optimal theory based on the LAN property. Meanwhile, the test based on the log-likelihood ratio is shown to be asymptotically most powerful with the benefit of the classical Neymann–Pearson framework.

Détail

Keywords: ANOVA, likelihood ratio process, local asymptotic normality, random effect, simultaneous equation.

Bibliographie

[1] R. L. Anderson, “Uses of variance component analysis in the interpretation of biological experiments”, Bull. Inst. Internat. Statist., 37:3 (1960), 71–90 | MR | Zbl

[2] T. W. Anderson, N. Kunitomo, K. Morimune, “Comparing single-equation estimators in a simultaneous equation system”, Econometric Theory, 2:1 (1986), 1–32 | DOI

[3] T. W. Anderson, H. Rubin, “Estimation of the parameters of a single equation in a complete system of stochastic equations”, Ann. Math. Statist., 20:1 (1949), 46–63 | DOI | MR | Zbl

[4] R. L. Basmann, “A generalized classical method of linear estimation of coefficients in a structural equation”, Econometrica, 25 (1957), 77–83 | DOI | MR | Zbl

[5] Y. Benghabrit, M. Hallin, “Locally asymptotically optimal tests for autoregressive against bilinear serial dependence”, Statist. Sinica, 6:1 (1996), 147–169 | MR | Zbl

[6] C. Cutting, D. Paindaveine, T. Verdebout, “Testing uniformity on high-dimensional spheres against monotone rotationally symmetric alternatives”, Ann. Statist., 45:3 (2017), 1024–1058 | DOI | MR | Zbl

[7] R. Dahlhaus, “Maximum likelihood estimation and model selection for locally stationary processes”, J. Nonparametr. Statist., 6:2-3 (1996), 171–191 | DOI | MR | Zbl

[8] G. E. Dickerson, “Experimental design for testing inbred lines of swine”, J. Animal Sci., 1:4 (1942), 326–341 | DOI

[9] R. A. Fisher, “XV.-The correlation between relatives on the supposition of Mendelian inheritance”, Trans. Roy. Soc. Edinburgh, 52:2 (1919), 399–433 | DOI

[10] Y. Fujikoshi, K. Morimune, N. Kunitomo, M. Taniguchi, “Asymptotic expansions of the distributions of the estimates of coefficients in a simultaneous equation system”, J. Econometrics, 18:2 (1982), 191–205 | DOI | MR | Zbl

[11] M. Hallin, J.-F. Ingenbleek, M. L. Puri, “Linear serial rank tests for randomness against ARMA alternatives”, Ann. Statist., 13:3 (1985), 1156–1181 | DOI | MR | Zbl

[12] M. Hallin, M. Taniguchi, A. Serroukh, K. Choy, “Local asymptotic normality for regression models with long-memory disturbance”, Ann. Statist., 27:6 (1999), 2054–2080 | DOI | MR | Zbl

[13] J. Hirukawa, M. Taniguchi, “LAN theorem for non-Gaussian locally stationary processes and its applications”, J. Statist. Plann. Inference, 136:3 (2006), 640–688 | DOI | MR | Zbl

[14] Y. Hosoya, Y. Tsukuda, N. Terui, “Ancillarity and the limited information maximum-likelihood estimation of a structural equation in a simultaneous equation system”, Econometric Theory, 5:3 (1989), 385–404 | DOI | MR

[15] I. A. Ibragimov, R. Z. Khas'minskii, “Local asymptotic normality for non-identically distributed observations”, Theory Probab. Appl., 20:2 (1976), 246–260 | DOI | MR | Zbl

[16] I. A. Ibragimov, R. Z. Has'minskii, Statistical estimation. Asymptotic theory, Appl. Math., 16, Springer-Verlag, New York–Berlin, 1981, vii+403 pp. | DOI | MR | MR | Zbl | Zbl

[17] P. Jeganathan, “Some aspects of asymptotic theory with applications to time series models”, Econometric Theory, 11:5 (1995), 818–887 | DOI | MR

[18] H. Kato, M. Taniguchi, M. Honda, “Statistical analysis for multiplicatively modulated nonlinear autoregressive model and its applications to electrophysiological signal analysis in humans”, IEEE Trans. Signal Process., 54:9 (2006), 3414–3425 | DOI | Zbl

[19] J.-P. Kreiss, “On adaptive estimation in stationary ARMA processes”, Ann. Statist., 15:1 (1987), 112–133 | DOI | MR | Zbl

[20] J.-P. Kreiss, “Local asymptotic normality for autoregression with infinite order”, J. Statist. Plann. Inference, 26:2 (1990), 185–219 | DOI | MR | Zbl

[21] L. Le Cam, “Locally asymptotically normal families of distributions. Certain approximations to families of distributions and their use in the theory of estimation and testing hypotheses”, Univ. California Publ. Statist., 3 (1960), 37–98 | MR | Zbl

[22] E. L. Lehmann, J. P. Romano, Testing statistical hypotheses, Springer Texts Statist., 3rd ed., Springer, New York, 2005, xiv+784 pp. | DOI | MR | MR | Zbl | Zbl

[23] O. Linton, “Adaptive estimation in ARCH models”, Econometric Theory, 9:4 (1993), 539–569 | DOI | MR

[24] D. Paindaveine, T. Verdebout, “Inference on the mode of weak directional signals: a Le Cam perspective on hypothesis testing near singularities”, Ann. Statist., 45:2 (2017), 800–832 | DOI | MR | Zbl

[25] P. C. B. Phillips, “Partially identified econometric models”, Econometric Theory, 5:2 (1989), 181–240 | DOI | MR

[26] G. G. Roussas, Contiguity of probability measures. Some applications in statistics, Cambridge Tracts in Math. and Math. Phys., 63, Cambridge Univ. Press, London–New York, 1972, xiii+248 pp. | DOI | MR | Zbl

[27] G. G. Roussas, “Asymptotic distribution of the log-likelihood function for stochastic processes”, Z. Wahrsch. Verw. Gebiete, 47:1 (1979), 31–46 | DOI | MR | Zbl

[28] K. Sakiyama, M. Taniguchi, “Testing composite hypotheses for locally stationary processes”, J. Time Ser. Anal., 24:4 (2003), 483–504 | DOI | MR | Zbl

[29] S. R. Searle, G. Casella, C. E. McCulloch, Variance components, Wiley Ser. Probab. Math. Statist. Appl. Probab. Statist., John Wiley Sons, Inc., New York, 1992, xxvi+501 pp. | DOI | MR | Zbl

[30] R. H. Sharpe, C. H. Van Middelem, “Application of variance components to horticultural problems with special reference to a parathion residue study”, Proc. Amer. Soc. Horticultural Sci., 66 (1955), 415–420

[31] G. F. Sprague, L. A. Tatum, “General vs. specific combining ability in single crosses of corn”, Agron. J., 34:10 (1942), 923–932 | DOI

[32] A. R. Swensen, “The asymptotic distribution of the likelihood ratio for autoregressive time series with a regression trend”, J. Multivariate Anal., 16:1 (1985), 54–70 | DOI | MR | Zbl

[33] K. Takeuchi, K. Morimune, “Third-order efficiency of the extended maximum likelihood estimators in a simultaneous equation system”, Econometrica, 53:1 (1985), 177–200 | DOI | MR | Zbl

[34] M. Talbot, “Yield variability of crop varieties in the U.K.”, J. Agric. Sci., 102:2 (1984), 315–321 | DOI

[35] M. Taniguchi, Y. Kakizawa, Asymptotic theory of statistical inference for time series, Springer Ser. Statist., Springer-Verlag, New York, 2000, xviii+661 pp. | DOI | MR | Zbl

[36] H. Theil, Repeated least squares applied to complete equation systems, Central planning bureau, The Hague, 1953

[37] H. Theil, Economic forecasts and policy, Contrib. Econom. Anal., 15, 2nd ed., North-Holland Publishing Co., Amsterdam, 1961, xxxii+567 pp.

[38] A. W. van der Vaart, Asymptotic statistics, Camb. Ser. Stat. Probab. Math., 3, paperback ed., Cambridge Univ. Press, Cambridge, 2000, xvi+443 pp. | DOI | MR | Zbl