## Bernoulli

• Bernoulli
• Volume 21, Number 2 (2015), 697-739.

### Testing equality of spectral densities using randomization techniques

#### Abstract

In this paper, we investigate the testing problem that the spectral density matrices of several, not necessarily independent, stationary processes are equal. Based on an $L_{2}$-type test statistic, we propose a new nonparametric approach, where the critical values of the tests are calculated with the help of randomization methods. We analyze asymptotic exactness and consistency of these randomization tests and show in simulation studies that the new procedures posses very good size and power characteristics.

#### Article information

Source
Bernoulli, Volume 21, Number 2 (2015), 697-739.

Dates
First available in Project Euclid: 21 April 2015

https://projecteuclid.org/euclid.bj/1429624958

Digital Object Identifier
doi:10.3150/13-BEJ584

Mathematical Reviews number (MathSciNet)
MR3338644

Zentralblatt MATH identifier
1320.62081

#### Citation

Jentsch, Carsten; Pauly, Markus. Testing equality of spectral densities using randomization techniques. Bernoulli 21 (2015), no. 2, 697--739. doi:10.3150/13-BEJ584. https://projecteuclid.org/euclid.bj/1429624958

#### References

• [1] Basrak, B., Davis, R.A. and Mikosch, T. (2002). Regular variation of GARCH processes. Stochastic Process. Appl. 99 95–115.
• [2] Beltrão, K.I. and Bloomfield, P. (1987). Determining the bandwidth of a kernel spectrum estimate. J. Time Ser. Anal. 8 21–38.
• [3] Brillinger, D.R. (1981). Time Series: Data Analysis and Theory, 2nd ed. Holden-Day Series in Time Series Analysis. Oakland, CA: Holden-Day.
• [4] Caiado, J., Crato, N. and Peña, D. (2006). A periodogram-based metric for time series classification. Comput. Statist. Data Anal. 50 2668–2684.
• [5] Caiado, J., Crato, N. and Peña, D. (2009). Comparison of times series with unequal length in the frequency domain. Comm. Statist. Simulation Comput. 38 527–540.
• [6] Coates, D.S. and Diggle, P.J. (1986). Tests for comparing two estimated spectral densities. J. Time Ser. Anal. 7 7–20.
• [7] de Jong, P. (1987). A central limit theorem for generalized quadratic forms. Probab. Theory Related Fields 75 261–277.
• [8] Dette, H. and Hildebrandt, T. (2012). A note on testing hypotheses for stationary processes in the frequency domain. J. Multivariate Anal. 104 101–114.
• [9] Dette, H. and Paparoditis, E. (2009). Bootstrapping frequency domain tests in multivariate time series with an application to comparing spectral densities. J. R. Stat. Soc. Ser. B Stat. Methodol. 71 831–857.
• [10] Diggle, P.J. and Fisher, N.I. (1991). Nonparametric comparison of cumulative periodograms. J. Roy. Statist. Soc. Ser. C 40 423–434.
• [11] Eichler, M. (2008). Testing nonparametric and semiparametric hypotheses in vector stationary processes. J. Multivariate Anal. 99 968–1009.
• [12] Hannan, E.J. (1970). Multiple Time Series. New York: Wiley.
• [13] Härdle, W. and Mammen, E. (1993). Comparing nonparametric versus parametric regression fits. Ann. Statist. 21 1926–1947.
• [14] Janssen, A. and Pauls, T. (2003). How do bootstrap and permutation tests work? Ann. Statist. 31 768–806.
• [15] Jentsch, C. (2012). A new frequency domain approach of testing for covariance stationarity and for periodic stationarity in multivariate linear processes. J. Time Series Anal. 33 177–192.
• [16] Jentsch, C. and Pauly, M. (2012). A note on using periodogram-based distances for comparing spectral densities. Statist. Probab. Lett. 82 158–164.
• [17] Jentsch, C. and Pauly, M. (2014). Supplement to “Testing equality of spectral densities using randomization techniques”. DOI:10.3150/13-BEJ584SUPP.
• [18] Lehmann, E.L. and Romano, J.P. (2005). Testing Statistical Hypotheses, 3rd ed. Springer Texts in Statistics. New York: Springer.
• [19] Liu, W. and Wu, W.B. (2010). Asymptotics of spectral density estimates. Econometric Theory 26 1218–1245.
• [20] Maharaj, E.A. (2002). Comparison of non-stationary time series in the frequency domain. Comput. Statist. Data Anal. 40 131–141.
• [21] Paparoditis, E. (2000). Spectral density based goodness-of-fit tests for time series models. Scand. J. Statist. 27 143–176.
• [22] Paparoditis, E. (2005). Testing the fit of a vector autoregressive moving average model. J. Time Ser. Anal. 26 543–568.
• [23] Paparoditis, E. (2009). Testing temporal constancy of the spectral structure of a time series. Bernoulli 15 1190–1221.
• [24] Pötscher, B.M. and Reschenhofer, E. (1988). Discriminating between two spectral densities in case of replicated observations. J. Time Ser. Anal. 9 221–224.
• [25] Preuß, P. and Hildebrandt, T. (2013). Comparing spectral densities of stationary time series with unequal sample sizes. Statist. Probab. Lett. 83 1174–1183.
• [26] Preuß, P., Vetter, M. and Dette, H. (2013). A test for stationarity based on empirical processes. Bernoulli 19 2715–2749.
• [27] Priestley, M.B. (1981). Spectral Analysis and Time Series. Vol. 2. Multivariate Series, Prediction and Control, Probability and Mathematical Statistics. London: Academic Press [Harcourt Brace Jovanovich Publishers].
• [28] Robinson, P.M. (1991). Automatic frequency domain inference on semiparametric and nonparametric models. Econometrica 59 1329–1363.
• [29] Taniguchi, M. and Kakizawa, Y. (2000). Asymptotic Theory of Statistical Inference for Time Series. Springer Series in Statistics. New York: Springer.
• [30] Taniguchi, M. and Kondo, M. (1993). Nonparametric approach in time series analysis. J. Time Ser. Anal. 14 397–408.
• [31] Taniguchi, M., Puri, M.L. and Kondo, M. (1996). Nonparametric approach for non-Gaussian vector stationary processes. J. Multivariate Anal. 56 259–283.
• [32] Wu, W.B. (2005). Nonlinear system theory: Another look at dependence. Proc. Natl. Acad. Sci. USA 102 14150–14154 (electronic).
• [33] Xiao, H. and Wu, W. (2014). Portmanteau test and simultaneous inference for serial covariances. Statist. Sinica. To appear. With supplementary material in http://arxiv.org/pdf/1105.3423.pdf. Preprint SS-11-212R2.

#### Supplemental materials

• Supplementary material: Supplement to “Testing equality of spectral densities using randomization techniques”. In the supplement to the current paper (cf. Jentsch and Pauly [17]), we provide additional supporting simulations for the asymptotic test and all three randomization tests under consideration in a variety of examples.