Abstract
Minimum discrimination information provides a useful generalization of likelihood methodology for classification and clustering of multivariate time series. Discrimination between different classes of multivariate time series that can be characterized by differing covariance or spectral structures is of importance in applications occurring in the analysis of geophysical and medical time series data. For discrimination between such multivariate series, Kullback-Leibler discrimination information and the Chernoff information measure are developed for the multivariate non-Gaussian case. Asymptotic error rates and limiting distributions are given for a generalized spectral disparity measure that includes the foregoing criteria as special cases. Applications to problems of clustering and classifying earthquakes and mining explosions are given.
| Original language | English |
|---|---|
| Pages (from-to) | 328-340 |
| Number of pages | 13 |
| Journal | Journal of the American Statistical Association |
| Volume | 93 |
| Issue number | 441 |
| Publication status | Published - 1998 Mar |
| Externally published | Yes |
Fingerprint
Keywords
- Chernoff
- Divergence
- Kullback-Leibler
- Minimum discrimination information
- Robustness
- Seismology
- Spectral analysis
ASJC Scopus subject areas
- Mathematics(all)
- Statistics and Probability
Cite this
Discrimination and clustering for multivariate time series. / Kakizawa, Yoshihide; Shumway, Robert H.; Taniguchi, Masanobu.
In: Journal of the American Statistical Association, Vol. 93, No. 441, 03.1998, p. 328-340.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Discrimination and clustering for multivariate time series
AU - Kakizawa, Yoshihide
AU - Shumway, Robert H.
AU - Taniguchi, Masanobu
PY - 1998/3
Y1 - 1998/3
N2 - Minimum discrimination information provides a useful generalization of likelihood methodology for classification and clustering of multivariate time series. Discrimination between different classes of multivariate time series that can be characterized by differing covariance or spectral structures is of importance in applications occurring in the analysis of geophysical and medical time series data. For discrimination between such multivariate series, Kullback-Leibler discrimination information and the Chernoff information measure are developed for the multivariate non-Gaussian case. Asymptotic error rates and limiting distributions are given for a generalized spectral disparity measure that includes the foregoing criteria as special cases. Applications to problems of clustering and classifying earthquakes and mining explosions are given.
AB - Minimum discrimination information provides a useful generalization of likelihood methodology for classification and clustering of multivariate time series. Discrimination between different classes of multivariate time series that can be characterized by differing covariance or spectral structures is of importance in applications occurring in the analysis of geophysical and medical time series data. For discrimination between such multivariate series, Kullback-Leibler discrimination information and the Chernoff information measure are developed for the multivariate non-Gaussian case. Asymptotic error rates and limiting distributions are given for a generalized spectral disparity measure that includes the foregoing criteria as special cases. Applications to problems of clustering and classifying earthquakes and mining explosions are given.
KW - Chernoff
KW - Divergence
KW - Kullback-Leibler
KW - Minimum discrimination information
KW - Robustness
KW - Seismology
KW - Spectral analysis
UR - http://www.scopus.com/inward/record.url?scp=0032343819&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0032343819&partnerID=8YFLogxK
M3 - Article
VL - 93
SP - 328
EP - 340
JO - Journal of the American Statistical Association
JF - Journal of the American Statistical Association
SN - 0162-1459
IS - 441
ER -