On the unique representation of non-Gaussian multivariate linear processes

@inproceedings{Chan2004OnTU,
  title={On the unique representation of non-Gaussian multivariate linear processes},
  author={Kung-Sik Chan and L M C Ho},
  year={2004}
}
In contrast to the fact that Gaussian linear processes generally have nonunique moving-average representations, non-Gaussian univariate linear processes have been shown to admit essentially unique moving-average representation, under various regularity conditions. We extend the one-dimensional result to multivariate processes. Under various conditions on the intercomponent dependence structure of the error process, we prove that for non-Gaussian multivariate linear processes the moving-average… CONTINUE READING

From This Paper

Topics from this paper.

References

Publications referenced by this paper.
Showing 1-10 of 11 references

Time Series: Theory and Methods

  • P. J. Brockwell, R. A. Davis
  • 1991
Highly Influential
1 Excerpt

Characterization Problems in Mathematical Statistics

  • A. M. Kagan, Y. V. Linnik, C. R. Rao
  • 1973
Highly Influential
2 Excerpts

On the unique representation of non-Gaussian linear processes

  • Q. Cheng
  • Ann. Statist.,
  • 1992
2 Excerpts

Amendments and corrections: The uniqueness of moving average representation with independent and identically distributed random variables for non-Gaussian stationary time series

  • D. Findley
  • Biometrika
  • 1990
2 Excerpts

Maximum standardized cumulant deconvolution of non-Gaussian linear processes

  • Q. Cheng
  • Ann. Statist.,
  • 1990

The uniqueness of moving average representation with independent and identically distributed random variables for non- Gaussian stationary time series

  • D. Findley
  • Biometrika
  • 1986
3 Excerpts

Stationary Sequences and Random Fields

  • M. Rosenblatt
  • 1985
1 Excerpt

Deconvolution and estimation of transfer function phase and coefficients for non-Gaussian linear processes

  • K. S. Lii, M. Rosenblatt
  • Ann. Statist.,
  • 1982
2 Excerpts

Multiple Time Series

  • E. J. Hannan
  • 1970
1 Excerpt

Similar Papers

Loading similar papers…