×

Affine-invariant aligned rank tests for the multivariate general linear model with VARMA errors. (English) Zbl 1087.62098

Summary: We develop optimal rank-based procedures for testing affine-invariant linear hypotheses on the parameters of a multivariate general linear model with elliptical VARMA errors. We propose a class of optimal procedures that are based either on residual (pseudo-) Mahalanobis signs and ranks, or on absolute interdirections and lift-interdirection ranks, i.e., on hyperplane-based signs and ranks. The Mahalanobis versions of these procedures are strictly affine-invariant, while the hyperplane-based ones are asymptotically affine-invariant. Both versions generalize the univariate signed rank procedures proposed by M. Hallin and M. L. Puri [ibid. 50, No. 2, 175–237 (1994; Zbl 0805.62050)], and are locally asymptotically most stringent under correctly specified radial densities. Their AREs with respect to Gaussian procedures are shown to be convex linear combinations of the AREs obtained by M. Hallin and D. Paindaveine [Ann. Stat. 30, 1103–1133 (2002; Zbl 1101.62348); Bernoulli 8, 787–815 (2002; Zbl 1018.62046)] for the pure location and purely serial models, respectively. The resulting test statistics are provided under closed form for several important particular cases, including multivariate Durbin-Watson tests, VARMA order identification tests, etc. The key technical result is a multivariate asymptotic linearity result proved by M. Hallin and D. Paindaveine [Asymptotic linearity of serial and nonserial multivariate signed rank statistics. J. Stat. Plann. Inference 136, 1–32 (2006; Zbl 1082.62049)].

MSC:

62M10 Time series, auto-correlation, regression, etc. in statistics (GARCH)
62G10 Nonparametric hypothesis testing
62H15 Hypothesis testing in multivariate analysis
62J05 Linear regression; mixed models
62G35 Nonparametric robustness
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Brockwell, P. J.; Davis, R. A., Time Series: Theory and Methods (1987), Springer: Springer New York · Zbl 0673.62085
[2] Dunsmuir, W.; Hannan, E. J., Vector linear time series, Adv. Appl. Probab, 8, 339-364 (1976) · Zbl 0327.62055
[3] Deistler, M.; Dunsmuir, W.; Hannan, E. J., Vector linear time series: corrections and extensions, Adv. Appl. Probab, 10, 360-372 (1978) · Zbl 0385.62072
[4] Garel, B.; Hallin, M., Local asymptotic normality of multivariate ARMA processes with a linear trend, Ann. Inst. Statist. Math, 47, 551-579 (1995) · Zbl 0841.62076
[5] Garel, B.; Hallin, M., Rank-based AR order identification, J. Amer. Statist. Assoc, 94, 1357-1371 (1999) · Zbl 0994.62089
[6] Hallin, M., Non-stationary \(q\)-dependent processes and time-varying moving-average modelsinvertibility properties and the forecasting problem, Adv. Appl. Probab, 18, 170-210 (1986) · Zbl 0597.62096
[7] Hallin, M.; Ingenbleek, J.-Fr; Puri, M. L., Asymptotically most powerful rank tests for multivariate randomness against serial dependence, J. Multivar. Anal, 30, 34-71 (1989) · Zbl 0685.62047
[8] Hallin, M.; Paindaveine, D., Optimal tests for multivariate location based on interdirections and pseudo-Mahalanobis ranks, Ann. Statist, 30, 1103-1133 (2002) · Zbl 1101.62348
[9] Hallin, M.; Paindaveine, D., Optimal procedures based on interdirections and pseudo-Mahalanobis ranks for testing multivariate elliptic white noise against ARMA dependence, Bernoulli, 8, 787-816 (2002) · Zbl 1018.62046
[10] M. Hallin, D. Paindaveine, Multivariate signed ranks: Randles’ interdirections or Tyler’s angles? in: Y. Dodge (Ed.), Statistical Data Analysis Based on the L1-Norm and Related Methods, Birkhäuser, Basel, 2002, pp. 271-282.; M. Hallin, D. Paindaveine, Multivariate signed ranks: Randles’ interdirections or Tyler’s angles? in: Y. Dodge (Ed.), Statistical Data Analysis Based on the L1-Norm and Related Methods, Birkhäuser, Basel, 2002, pp. 271-282. · Zbl 1145.62339
[11] M. Hallin, D. Paindaveine, Affine invariant linear hypotheses for the multivariate general linear model with VARMA error terms, in M. Moore, S. Froda, Chr. Léger (Eds.), Mathematical Statistics and Applications: Festschrift for Constance van Eeden, I.M.S. Lecture Notes-Monograph Series, I.M.S., Hayward, CA, 2003, pp. 417-434.; M. Hallin, D. Paindaveine, Affine invariant linear hypotheses for the multivariate general linear model with VARMA error terms, in M. Moore, S. Froda, Chr. Léger (Eds.), Mathematical Statistics and Applications: Festschrift for Constance van Eeden, I.M.S. Lecture Notes-Monograph Series, I.M.S., Hayward, CA, 2003, pp. 417-434.
[12] M. Hallin, D. Paindaveine, Rank-based optimal tests of the adequacy of an elliptic VARMA model, Ann. Statist. (2004), to appear.; M. Hallin, D. Paindaveine, Rank-based optimal tests of the adequacy of an elliptic VARMA model, Ann. Statist. (2004), to appear. · Zbl 1076.62044
[13] M. Hallin, D. Paindaveine, Asymptotic linearity of serial and nonserial multivariate signed rank statistics, J. Statist. Plann. Inference (2004), to appear.; M. Hallin, D. Paindaveine, Asymptotic linearity of serial and nonserial multivariate signed rank statistics, J. Statist. Plann. Inference (2004), to appear. · Zbl 1082.62049
[14] Hallin, M.; Puri, M. L., Optimal rank-based procedures for time-series analysistesting an ARMA model against other ARMA models, Ann. Statist, 16, 402-432 (1988) · Zbl 0659.62111
[15] Hallin, M.; Puri, M. L., Time-series analysis via rank-order theorysigned-rank tests for ARMA models, J. Multivar. Anal, 39, 1-29 (1991) · Zbl 0751.62041
[16] Hallin, M.; Puri, M. L., Aligned rank tests for linear models with autocorrelated error terms, J. Multivar. Anal, 50, 175-237 (1994) · Zbl 0805.62050
[17] Hallin, M.; Puri, M. L., A multivariate Wald-Wolfowitz rank test against serial dependence, Canad. J. Statist, 23, 55-65 (1995) · Zbl 0821.62022
[18] M. Hallin, B.J.M. Werker, Optimal testing for semi-parametric AR models: from Gaussian Lagrange multipliers to autoregression rank scores and adaptive tests, in: S. Ghosh (Ed.), Asymptotics, Nonparametrics, and Time Series, Marcel Dekker, New York, 1999, pp. 295-350.; M. Hallin, B.J.M. Werker, Optimal testing for semi-parametric AR models: from Gaussian Lagrange multipliers to autoregression rank scores and adaptive tests, in: S. Ghosh (Ed.), Asymptotics, Nonparametrics, and Time Series, Marcel Dekker, New York, 1999, pp. 295-350. · Zbl 1069.62541
[19] Hallin, M.; Werker, B. J.M., Semiparametric efficiency, distribution-freeness, and invariance, Bernoulli, 9, 137-165 (2003) · Zbl 1020.62042
[20] Hettmansperger, T. P.; Nyblom, J.; Oja, H., Affine invariant multivariate one-sample sign tests, J. R. Statist. Soc. Ser. B, 56, 221-234 (1994) · Zbl 0795.62055
[21] Hettmansperger, T. P.; Möttönen, J.; Oja, H., Affine invariant multivariate one-sample signed-rank tests, J. Amer. Statist. Assoc, 92, 1591-1600 (1997) · Zbl 0943.62051
[22] Jan, S.-L.; Randles, R. H., A multivariate signed-sum test for the one-sample location problem, J. Nonparametric Statist, 4, 49-63 (1994) · Zbl 1380.62192
[23] Kreiss, J. P., On adaptive estimation in stationary ARMA processes, Ann. Statist, 15, 112-133 (1987) · Zbl 0616.62042
[24] Le Cam, L., Locally asymptotically normal families of distributions, Univ. Calif. Pub. Statist, 3, 37-98 (1960)
[25] Le Cam, L., Asymptotic Methods in Statistical Decision Theory (1986), Springer: Springer New York · Zbl 0605.62002
[26] J. Marden, Multivariate rank tests, in: S. Ghosh (Ed.), Multivariate Analysis, Design of Experiments, and Sampling Survey, M. Dekker, New York, 1999, pp. 401-432.; J. Marden, Multivariate rank tests, in: S. Ghosh (Ed.), Multivariate Analysis, Design of Experiments, and Sampling Survey, M. Dekker, New York, 1999, pp. 401-432. · Zbl 0946.62060
[27] Möttönen, J.; Oja, H., Multivariate spatial sign and rank methods, J. Nonparametric Statist, 5, 201-213 (1995) · Zbl 0857.62056
[28] Oja, H., Affine invariant multivariate sign and rank tests and corresponding estimatesa review, Scand. J. Statist, 26, 319-343 (1999) · Zbl 0938.62063
[29] H. Oja, D. Paindaveine, Optimal signed-rank tests based on hyperplanes, J. Statist. Plann. Inference (2004), to appear.; H. Oja, D. Paindaveine, Optimal signed-rank tests based on hyperplanes, J. Statist. Plann. Inference (2004), to appear. · Zbl 1162.62353
[30] Peters, D.; Randles, R. H., A multivariate signed-rank test for the one-sample location problem, J. Amer. Statist. Assoc, 85, 552-557 (1990) · Zbl 0709.62051
[31] Pötscher, B. M., Order estimation in ARMA models by Lagrangian multiplier tests, Ann. Statist, 11, 872-885 (1983) · Zbl 0521.62072
[32] Pötscher, B. M., The behaviour of the Lagrange multiplier test in testing the orders of an ARMA model, Metrika, 32, 129-150 (1985) · Zbl 0606.62101
[33] Puri, M. L.; Sen, P. K., Nonparametric Methods in Multivariate Analysis (1971), Wiley: Wiley New York
[34] Randles, R. H., A distribution-free multivariate sign test based on interdirections, J. Amer. Statist. Assoc, 84, 1045-1050 (1989) · Zbl 0702.62039
[35] Randles, R. H., A simpler, affine-invariant, multivariate, distribution-free sign test, J. Amer. Statist. Assoc, 95, 1263-1268 (2000) · Zbl 1009.62047
[36] Randles, R. H.; Um, Y., Nonparametric tests for the multivariate multi-sample location problem, Statist. Sin, 8, 801-812 (1998) · Zbl 0905.62048
[37] Rao, C. R.; Mitra, S. K., Generalized Inverses of Matrices and its Applications (1971), Wiley: Wiley New York
[38] Reinsel, G. C., Elements of multivariate Time Series Analysis (1997), Springer: Springer New York · Zbl 1208.62141
[39] Swensen, A. R., The asymptotic distribution of the likelihood ratio for autoregressive time series with a regression trend, J. Multivar. Anal, 16, 54-70 (1985) · Zbl 0563.62065
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. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.