Estimation with incomplete data: the linear case
Pages 5082 - 5088
Abstract
Traditional methods for handling incomplete data, including Multiple Imputation and Maximum Likelihood, require that the data be Missing At Random (MAR). In most cases, however, missingness in a variable depends on the underlying value of that variable. In this work, we devise model-based methods to consistently estimate mean, variance and covariance given data that are Missing Not At Random (MNAR). While previous work on MNAR data require variables to be discrete, we extend the analysis to continuous variables drawn from Gaussian distributions. We demonstrate the merits of our techniques by comparing it empirically to state of the art software packages.
References
[1]
Jon Adams. Researching Complementary and Alternative Medicine. Routledge, 2007.
[2]
Paul D. Allison. Missing data techniques for structural equation modeling. Journal of Abnormal Psychology, 112(4):545, 2003.
[3]
Carlos Brito. Graphical Models for Identification in Structural Equation Models. PhD thesis, University of California Los Angeles, 2004.
[4]
Rhian M. Daniel, Michael G. Kenward, Simon N. Cousens, and Bianca L. De Stavola. Using causal diagrams to guide analysis in missing data problems. Statistical Methods in Medical Research, 21(3):243- 256, 2012.
[5]
Adnan Darwiche. Modeling and Reasoning with Bayesian Networks. Cambridge University Press, 2009.
[6]
Craig K. Enders. Analyzing structural equation models with missing data. In Gregory R. Hancock and Ralph O. Mueller, editors, Structural Equation Modeling: A Second Course, pages 315-344. Information Age Publishing, Inc., 2006.
[7]
John W. Graham. Adding missing-data-relevant variables to FIML-based structural equation models. Structural Equation Modeling, 10(1):80-100, 2003.
[8]
Sander Greenland and William D. Finkle. A critical look at methods for handling missing covariates in epidemiologic regression analyses. American Journal of Epidemiology, 142(12):1255-1264, 1995.
[9]
Daphne Koller and Nir Friedman. Probabilistic Graphical Models: Principles and Techniques. The MIT Press, 2009.
[10]
Roderick J. A. Little. Modeling the dropout mechanism in repeated-measures studies. Journal of the American Statistical Association, 90(431):1112-1121, 1995.
[11]
Karthika Mohan and Judea Pearl. Graphical models for recovering probabilistic and causal queries from missing data. In C. Cortes M. Welling, Z. Ghahramani and N. Lawrence, editors, Advances in Neural Information Processing Systems 27, pages 1520- 1528. Curran Associates, 2014.
[12]
Karthika Mohan and Judea Pearl. On the testability of models with missing data. In S. Kaski and J. Corander, editors, Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics (AISTATS), volume 33, pages 643-650, 2014.
[13]
Karthika Mohan and Judea Pearl. Graphical models for processing missing data. arXiv preprint arXiv:1801.03583, 2018.
[14]
Karthika Mohan, Judea Pearl, and Jin Tian. Graphical models for inference with missing data. In C.J.C. Burges, L. Bottou, M. Welling, Z. Ghahramani, and K.Q. Weinberger, editors, Advances in Neural Information Processing Systems 26, pages 1277-1285. Curran Associates, 2013.
[15]
Jason W. Osborne. Best Practices in Logistic Regression. SAGE Publications, 2014.
[16]
Judea Pearl. Causality: Models, Reasoning and Inference. Cambridge Univ Press, New York, 2nd edition, 2009.
[17]
Judea Pearl. Linear models: A useful "microscope" for causal analysis. Journal of Causal Inference, 1(1):155-170, 2013.
[18]
Donald B. Rubin. Inference and missing data. Biometrika, 63:581-592, 1976.
[19]
Gabriel L. Schlomer, Sheri Bauman, and Noel A. Card. Best practices for missing data management in counseling psychology. Journal of Counseling Psychology, 57(1):1, 2010.
[20]
Donna L. Schminkey, Timo von Oertzen, and Linda Bullock. Handling missing data with multilevel structural equation modeling and full information maximum likelihood techniques. Research in Nursing & Health, 39(4):286-297, 2016.
[21]
Iyla Shpitser, Karthika Mohan, and Judea Pearl. Missing data as a causal and probabilistic problem. In M. Meila and T. Heskes, editors, Proceedings of the Thirty-First Conference on Uncertainty in Artificial Intelligence, pages 802-811. AUAI Press, 2015.
[22]
Oleksandr Sverdlov. Modern Adaptive Randomized Clinical Trials: Statistical and Practical Aspects. Chapman and Hall/CRC, 2015.
[23]
Felix Thoemmes and Karthika Mohan. Graphical representation of missing data problems. Structural Equation Modeling: A Multidisciplinary Journal, 22:631-642, 2015.
[24]
Jodie B. Ullman and Peter M. Bentler. Structural Equation Modeling. Wiley Online Library, 2003.
[25]
Bas van Stein and Wojtek Kowalczyk. An incremental algorithm for repairing training sets with missing values. In International Conference on Information Processing and Management of Uncertainty in Knowledge-Based Systems, pages 175-186. Springer, 2016.
[26]
Sewall Wright. Correlation and causation. Journal of Agricultural Research, 20(7):557-585, 1921.
Recommendations
Software Cost Estimation with Incomplete Data
The construction of software cost estimation models remains an active topic of research. The basic premise of cost modeling is that a historical database of software project cost data can be used to develop a quantitative model to predict the cost of ...
Missing data imputation by utilizing information within incomplete instances
This paper proposes to utilize information within incomplete instances (instances with missing values) when estimating missing values. Accordingly, a simple and efficient nonparametric iterative imputation algorithm, called the NIIA method, is designed ...
Mining Incomplete Data with Many Lost and Attribute-Concept Values
Rough Sets and Knowledge TechnologyAbstractThis paper presents experimental results on twelve data sets with many missing attribute values, interpreted as lost values and attribute-concept values. Data mining was accomplished using three kinds of probabilistic approximations: singleton, ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
Sponsors
- Adobe
- IBMR: IBM Research
- ERICSSON
- Microsoft: Microsoft
- AI Journal: AI Journal
Publisher
AAAI Press
Publication History
Published: 13 July 2018
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 11 Dec 2024