Abstract
We consider the problem of multivariate outlier testing for purposes of distinguishing seismic signals of underground nuclear events from training samples based on non-nuclear seismic events when certain data are missing. We consider the case in which the training data follow a multivariate normal distribution. Assume a potential outlier is observed on which k features of interest are measured. Assume further that the available training set of n observations on these k features is available but that some of the observations in the training data have missing features. The approach currently used in practice is to perform the outlier testing using a generalized likelihood ratio test procedure based only on the data vectors in the training data with complete data. When there is a substantial amount of missing data within the training set, use of this strategy may lead to a loss of valuable information. An alternative procedure is to incorporate all n of the data vectors in the training data using the EM algorithm to appropriately handle the missing data in the training set. Resampling methods are used to find appropriate critical regions. We use simulation results and analysis of models fit to Pg/Lg ratios for the WMQ station in China to compare these two strategies for dealing with missing data.
*
This research was partially supported by DTRA01-99-C-0018
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Dempster A. P., Laird N. M., and Rubin, D. B. (1977), Maximum Likelihood Estimation from Incomplete Data via the EM Algorithm (with discussion), J. Roy. Statist. Soc. B39, 1–38.
Efron, B. and Tibshirani, R. J. An Introduction to the Bootstrap. (Chapman and Hall, New York 1993).
Fisk M. D., Gray H. L., and McCartor, G (1996), Regional Event Discrimination Without Transporting Thresholds, Bull. Seismol. Soc. Am. 86, 1545–1558.
Hartse H. E., Taylor S. R., Phillips W. S., and Randall, G. E. (1997), A Preliminary Study of Regional Seismic Discrimination in Central Asia with Emphasis on Western China, Bull. Seismol. Soc. Am. 87, 551–568.
Johnson, R. A. and Wichern, D. W., Applied Multivariate Statistical Analysis, Fourth Edition (Upper Saddle River, New Jersey: Prentice Hall 1998).
Little, R. J. A. and Rubin, D. B., Statistical Analysis with Missing Data (John Wiley and Sons, Inc. New York 1987).
McLachlan, G. J. (1987), On Bootstrapping the Likelihood Ratio Test Statistic for the Number of Components in a Normal Mixture, Appl. Statist. 36, 318–324.
Miller J. W., Gray H. L., and Woodward, W. A. (1993), Discriminant Analysis and Outlier Testing when Data are Missing, Phillips Laboratory Technical Report, ARPA F29601-91-K-DB25.
Miller J. W., Woodward W. A., Gray H. L., Fisk M. A., and McCartor, (1994), A Hypothesistesting Approach to Discriminant Analysis with Mixed Categorical and Continuous Variables when Data are Missing, Technical Report No. SMU/DS/TR-273, Department of Statistical Science, Southern Methodist University.
Sain S. R., Gray H. L., Woodword W. A., and Fisk, M. D. (1999), Outlier Detection when Training Data are Unlabled, Bull. Seismol. Soc. Am. 89, 294–304.
Taylor, S. R. and Hartse, H. E. (1997), An Evaluation of Generalized Likelihood Ratio Outlier Detection to Identification of Seismic Events in Western China, Bull. Seismol. Soc. Am. 87, 824–831.
Wang S., Woodward W. A., Gray H. L., Wiechecki S., and Sain, S. R. (1997), A New test for Outlier Detection from a Multivariate Mixture Distribution, J. Computat. and Graph. Statist. 6, 285–299.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer Basel AG
About this chapter
Cite this chapter
Woodward, W.A., Sain, S.R., Gray, H.L., Zhao, B., Fisk, M.D. (2002). Testing for Multivariate Outliers in the Presence of Missing Data. In: Walter, W.R., Hartse, H.E. (eds) Monitoring the Comprehensive Nuclear-Test-Ban Treaty: Seismic Event Discrimination and Identification. Pageoph Topical Volumes. Birkhäuser, Basel. https://doi.org/10.1007/978-3-0348-8169-2_14
Download citation
DOI: https://doi.org/10.1007/978-3-0348-8169-2_14
Publisher Name: Birkhäuser, Basel
Print ISBN: 978-3-7643-6675-9
Online ISBN: 978-3-0348-8169-2
eBook Packages: Springer Book Archive