Correlation when data are missing
- 1 June 2010
- journal article
- Published by Taylor & Francis Ltd in Journal of the Operational Research Society
- Vol. 61 (6), 1049-1056
- https://doi.org/10.1057/jors.2009.49
Abstract
Variable correlation is important for many operations research models. Many inventory, revenue management, and queuing models presume uncorrelated demand between products, market segments, or time periods. The specific model applied, or the resulting policies of a model, can differ drastically depending on variable correlation. Having missing data are a common problem for the real world application of operations research models. This work is at the junction of the two topics of correlation and missing data. We propose a test of independence between two variables when data are missing. The typical method for determining correlation with missing data ignores all data pairs in which one point is missing. The test presented here incorporates all data. The test can be applied when both variables are continuous, when both are discrete, or when one variable is discrete and the other is continuous. The test makes no assumptions about the distribution of the two variables, and thus it can be used to extend application of non-parametric rank tests, such as Spearman's rank correlation, to the case where data are missing. An example is shown where failure to incorporate the incomplete data yields incorrect policies.Keywords
This publication has 16 references indexed in Scilit:
- Estimating bus passenger waiting times from incomplete bus arrivals dataJournal of the Operational Research Society, 2007
- Technology scoring model considering rejected applicants and effect of reject inferenceJournal of the Operational Research Society, 2007
- The comparative efficacy of imputation methods for missing data in structural equation modelingEuropean Journal of Operational Research, 2003
- Integrated multi-item production-inventory systemsEuropean Journal of Operational Research, 1996
- A resampling method based on pivotal estimating functionsBiometrika, 1994
- Centralized versus Decentralized Manpower Resource Planning: the Case of a Hong Kong CompanyJournal of the Operational Research Society, 1991
- Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated dataBiometrika, 1990
- Inference and missing dataBiometrika, 1976
- Inventory control of a multiproduct system with a limited production resourceNaval Research Logistics Quarterly, 1967
- Chi-Square Tests with One Degree of Freedom; Extensions of the Mantel-Haenszel ProcedureJournal of the American Statistical Association, 1963