Imputation for multisource data with comparison and assessment techniques
Abstract
Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state-space model. These methods, as well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank themethods,we also propose an approach to identify significant differences. Imputation techniqueswill also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k-means clustering. In general, we found that imputation through a dynamic linearmodel tended to be themore »
- Authors:
-
- Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
- Publication Date:
- Research Org.:
- Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1416279
- Report Number(s):
- LA-UR-17-23333
Journal ID: ISSN 1524-1904
- Grant/Contract Number:
- AC52-06NA25396
- Resource Type:
- Journal Article: Accepted Manuscript
- Journal Name:
- Applied Stochastic Models in Business and Industry
- Additional Journal Information:
- Journal Volume: 34; Journal Issue: 1; Journal ID: ISSN 1524-1904
- Publisher:
- Wiley
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; Imputation; assessment; multisource data; dynamic linear model; comparison; k-means clustering
Citation Formats
Casleton, Emily Michele, Osthus, David Allen, and Van Buren, Kendra Lu. Imputation for multisource data with comparison and assessment techniques. United States: N. p., 2017.
Web. doi:10.1002/asmb.2299.
Casleton, Emily Michele, Osthus, David Allen, & Van Buren, Kendra Lu. Imputation for multisource data with comparison and assessment techniques. United States. https://doi.org/10.1002/asmb.2299
Casleton, Emily Michele, Osthus, David Allen, and Van Buren, Kendra Lu. 2017.
"Imputation for multisource data with comparison and assessment techniques". United States. https://doi.org/10.1002/asmb.2299. https://www.osti.gov/servlets/purl/1416279.
@article{osti_1416279,
title = {Imputation for multisource data with comparison and assessment techniques},
author = {Casleton, Emily Michele and Osthus, David Allen and Van Buren, Kendra Lu},
abstractNote = {Missing data are prevalent issue in analyses involving data collection. The problem of missing data is exacerbated for multisource analysis, where data from multiple sensors are combined to arrive at a single conclusion. In this scenario, it is more likely to occur and can lead to discarding a large amount of data collected; however, the information from observed sensors can be leveraged to estimate those values not observed. We propose two methods for imputation of multisource data, both of which take advantage of potential correlation between data from different sensors, through ridge regression and a state-space model. These methods, as well as the common median imputation, are applied to data collected from a variety of sensors monitoring an experimental facility. Performance of imputation methods is compared with the mean absolute deviation; however, rather than using this metric to solely rank themethods,we also propose an approach to identify significant differences. Imputation techniqueswill also be assessed by their ability to produce appropriate confidence intervals, through coverage and length, around the imputed values. Finally, performance of imputed datasets is compared with a marginalized dataset through a weighted k-means clustering. In general, we found that imputation through a dynamic linearmodel tended to be the most accurate and to produce the most precise confidence intervals, and that imputing the missing values and down weighting them with respect to observed values in the analysis led to the most accurate performance.},
doi = {10.1002/asmb.2299},
url = {https://www.osti.gov/biblio/1416279},
journal = {Applied Stochastic Models in Business and Industry},
issn = {1524-1904},
number = 1,
volume = 34,
place = {United States},
year = {Wed Dec 27 00:00:00 EST 2017},
month = {Wed Dec 27 00:00:00 EST 2017}
}
Web of Science
Figures / Tables:
Works referenced in this record:
Foci and Bases of Employee Commitment: Implications for job Performance.
journal, April 1996
- Becker, T. E.; Billings, R. S.; Eveleth, D. M.
- Academy of Management Journal, Vol. 39, Issue 2
An Entropy Weighting k-Means Algorithm for Subspace Clustering of High-Dimensional Sparse Data
journal, August 2007
- Jing, Liping; Ng, Michael K.; Huang, Joshua Zhexue
- IEEE Transactions on Knowledge and Data Engineering, Vol. 19, Issue 8
Methods for imputation of missing values in air quality data sets
journal, June 2004
- Junninen, Heikki; Niska, Harri; Tuppurainen, Kari
- Atmospheric Environment, Vol. 38, Issue 18
Missing value estimation methods for DNA microarrays
journal, June 2001
- Troyanskaya, O.; Cantor, M.; Sherlock, G.
- Bioinformatics, Vol. 17, Issue 6
Ridge Regression: Biased Estimation for Nonorthogonal Problems
journal, February 1970
- Hoerl, Arthur E.; Kennard, Robert W.
- Technometrics, Vol. 12, Issue 1
A Bayesian Study of the Error-in-Variables Model
journal, August 1981
- Reilly, Park M.; Patino-Leal, Hugo
- Technometrics, Vol. 23, Issue 3
An analysis of four missing data treatment methods for supervised learning
journal, May 2003
- Batista, Gustavo E. A. P. A.; Monard, Maria Carolina
- Applied Artificial Intelligence, Vol. 17, Issue 5-6
Bi-level multi-source learning for heterogeneous block-wise missing data
journal, November 2014
- Xiang, Shuo; Yuan, Lei; Fan, Wei
- NeuroImage, Vol. 102
Regularization Paths for Generalized Linear Models via Coordinate Descent
journal, January 2010
- Friedman, Jerome; Hastie, Trevor; Tibshirani, Robert
- Journal of Statistical Software, Vol. 33, Issue 1
Sequential Imputations and Bayesian Missing Data Problems
journal, March 1994
- Kong, Augustine; Liu, Jun S.; Wong, Wing Hung
- Journal of the American Statistical Association, Vol. 89, Issue 425
Spectral methods for imputation of missing air quality data
journal, December 2015
- Moshenberg, Shai; Lerner, Uri; Fishbain, Barak
- Environmental Systems Research, Vol. 4, Issue 1
An Introduction to Statistical Learning
book, January 2013
- James, Gareth; Witten, Daniela; Hastie, Trevor
- Springer Texts in Statistics
Review: A gentle introduction to imputation of missing values
journal, October 2006
- Donders, A. Rogier T.; van der Heijden, Geert J. M. G.; Stijnen, Theo
- Journal of Clinical Epidemiology, Vol. 59, Issue 10
The Calculation of Posterior Distributions by Data Augmentation
journal, June 1987
- Tanner, Martin A.; Wong, Wing Hung
- Journal of the American Statistical Association, Vol. 82, Issue 398
Sensor fusion potential exploitation-innovative architectures and illustrative applications
journal, January 1997
- Dasarathy, B. V.
- Proceedings of the IEEE, Vol. 85, Issue 1
An introduction to multisensor data fusion
journal, January 1997
- Hall, D. L.; Llinas, J.
- Proceedings of the IEEE, Vol. 85, Issue 1
Figures / Tables found in this record: