Electronic Journal of Statistics

Structural shrinkage of nonparametric spectral estimators for multivariate time series

Hilmar Böhm and Rainer von Sachs

Full-text: Open access

Abstract

In this paper we investigate the performance of periodogram based estimators of the spectral density matrix of possibly high-dimensional time series. We suggest and study shrinkage as a remedy against numerical instabilities due to deteriorating condition numbers of (kernel) smoothed periodogram matrices. Moreover, shrinking the empirical eigenvalues in the frequency domain towards one another also improves at the same time the Mean Squared Error (MSE) of these widely used nonparametric spectral estimators. Compared to some existing time domain approaches, restricted to i.i.d. data, in the frequency domain it is necessary to take the size of the smoothing span as “effective or local sample size” into account. While Böhm & von Sachs (2007) proposes a multiple of the identity matrix as optimal shrinkage target in the absence of knowledge about the multidimensional structure of the data, here we consider “structural” shrinkage. We assume that the spectral structure of the data is induced by underlying factors. However, in contrast to actual factor modelling suffering from the need to choose the number of factors, we suggest a model-free approach. Our final estimator is the asymptotically MSE-optimal linear combination of the smoothed periodogram and the parametric estimator based on an underfitting (and hence deliberately misspecified) factor model. We complete our theoretical considerations by some extensive simulation studies. In the situation of data generated from a higher-order factor model, we compare all four types of involved estimators (including the one of Böhm & von Sachs (2007)).

Article information

Source
Electron. J. Statist., Volume 2 (2008), 696-721.

Dates
First available in Project Euclid: 13 August 2008

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1218631741

Digital Object Identifier
doi:10.1214/08-EJS236

Mathematical Reviews number (MathSciNet)
MR2430251

Zentralblatt MATH identifier
1320.62198

Citation

Böhm, Hilmar; von Sachs, Rainer. Structural shrinkage of nonparametric spectral estimators for multivariate time series. Electron. J. Statist. 2 (2008), 696--721. doi:10.1214/08-EJS236. https://projecteuclid.org/euclid.ejs/1218631741


Export citation

References

  • Bai, J. & Ng, S. (2002), ‘Determining the number of factors in approximate factor models’, Econometrica 70(1), 191–221.
  • Beran, R. & Dümbgen, L. ( 1998), ‘Modulation of estimators and confidence sets’, The Annals of Statistics 26(5), 1826–1856.
  • Bickel, P. & Levina, E. (2007), Covariance regularization by thresholding., The Annals of Statistics, to appear.
  • Böhm, H. & von Sachs, R. (2007), ‘Shrinkage estimation in the frequency domain of multivariate time series’, Discussion paper 0706, Institut de Statistique, Université Catholique de, Louvain.
  • Botts, C. & Daniels, M. (2006), ‘A shrinkage estimator of spectral densities’, Biometrika 93, 179–195.
  • Brillinger, D. R. (1975), Time Series. Data Analysis and Theory, Holt, Reinhart and Winston.
  • Brockwell, P. & Davis, R. (1987), Time Series: Theory and Methods, Springer.
  • Daniels, M. & Cressie, N. (2001), ‘A hierarchical approach to covariance function estimation for time series’, Journal of Time Series Analysis 22, 253–266.
  • Forni, M., Hallin, M., Lippi, M. & Reichlin, L. ( 2000), ‘The generalized dynamic-factor model: Identification and estimation’, The Review of Economics and Statistics 82(4), 540–554.
  • Haff, L. (1977), ‘Minimax estimators for a multinormal precision matrix’, Journal of Multivariate Analysis 7, 374–385.
  • Haff, L. (1979), ‘Estimation of the inverse covariance matrix: Random mixtures of the inverse wishart matrix and the identity’, The Annals of Statistics 7(6), 1264–1276.
  • Haff, L. (1980), ‘Empirical bayes estimation of the multivariate normal covariance matrix’, The Annals of Statistics 8, 586–597.
  • Hallin, M. & Liška, R. (2007), ‘Determining the number of factors in the general dynamic factor model’, Journal of the American Statistical Association 102(478), 603–617.
  • Hoerl, A. & Kennard, R. (1970), ‘Ridge regression: biased estimation for nonorthogonal problems’, Technometrics 12(1), 55–67.
  • Jolliffe, I. (2002), Principal Component Analysis, Springer.
  • Kakizawa, Y., Shumway, R. & Taniguchi, M. ( 1998), ‘Discrimination and clustering for multivariate time series’, Journal of the American Statistical Association 93(441), 328–340.
  • Kullback, S. & Leibler, R. (1952), ‘On information and sufficiency’, Annals of Mathematical Statistics 22(1), 79–86.
  • Ledoit, O. & Wolf, M. (2003), ‘Improved estimation of the covariance matrix of stock returns with an application to portfolio selection’, Journal of Empirical Finance 10, 603–621.
  • Ledoit, O. & Wolf, M. (2004), ‘A well-conditioned estimator for large-dimensional covariance matrices’, Journal of Multivariate Analysis 88, 365–411.
  • Rothman, A., Levina, E. & Zhu, J. ( 2008), Generalized thresholding of large covariance matrices. Technical report 477, Dept. of Statistics, Univ. of, Michigan.
  • Sachs, L. & Hedderich, J. (2006), Angewandte Statistik, Springer.
  • Sharpe, W. (1963), ‘A simplified model for portfolio analysis’, Management Science 9, 277–293.
  • Shumway, R. H. & Stoffer, D. S. ( 2000), Time Series Analysis and its Applications, Springer.