Abstract
Statistical pattern recognition has attracted great interest due to its applicability and to the advances in technology and computing. Significant research has been done in areas such as automatic character recognition, medical diagnostics, and data mining. The classical discrimination rule for pattern recognition assumes normality. However, in real life this assumption is often questionable. In some situations, the pattern vector is a mixture of discrete and continuous random variables. In this article, we use copula densities to model class conditional distribution for pattern recognition with Bayes’ decision rule. These types of densities are useful when the marginal densities of a pattern vector are not normally distributed. Those models are also useful for mixed pattern vectors. We also did simulations to compare the performance of the copula-based classifier with classical normal distribution based model and the independent-assumption-based model.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Aeberhard, S., D. Coomans, and O. de Vel. 1994. Comparative analysis of statistical pattern recognition methods in high dimensional settings. Pattern Recog., 27(8), 1065–1077.
Czado, C., R. Kastenmeier, E. C. Brechmann, and A. Min. 2012. A mixed copula model for insurance claims and claim sizes. Scand. Actuarial J., 2012(4), 278–305.
Efron, B. 1975. The efficiency of logistic regression compared to normal discriminant analysis. J. Am. Stat. Assoc., 70, 892–898.
Friedman, J. H. 1989. Regularized discriminant analysis. J. Am. Stat. Assoc., 84, 165–175.
Gupta, A. K., ed. 1987. Error rate estimation in discriminant analysis: Recent advances, 233–252. Dordrecht, The Netherlands: Reidal.
James, G. M., and T. J. Hastie. 2001. Functional linear discriminant analysis for irregularly samples curves. J. R. Stat. Soc. Ser. B, 63(3), 533–550.
Joe, H. 2005. Asymptotic efficiency of the two-stage estimation method for copula-based models. J. Multivariate Anal., 94, 401–419.
Joe, H. 2014. Dependence modeling with copulas. Monographs on Statistics & Applied Probability. Boca Raton, FL: Chapman & Hall/CRC.
Krzanowski, W. J., P. Jonathan, W. V. McCarthy, and M. R. Thomas. 1995. Discriminant analysis with singular covariance matrices: Methods and applications to spectroscopic data. App. Stat., 44(1), 101–115.
Nelson, R. B. 2006. An introduction to copulas. Springer Series in Statistics. New York, NY: Springer.
Pohar, M., M. Blas, and S. Turk. 2004. Comparison of logistic regression and linear discriminant analysis. Metodoloki Zvezki, 1(1), 143–161.
Salinas-Gutierrez, R., A. Hernandez-Aguirre, M. J. J. Rivera-Meraz, and E. R. Villa-Diharce. 2011. Using Gaussian copulas in supervised probabilistic classification: Soft computing for intelligent control and mobile robotics studies. Comput. Intell., 318, 355–372.
Song, P. X. K. 2007. Correlated data analysis: Modeling, analytics, and applications. Springer Series in Statistics, vol. 365. New York, NY: Springer.
Song, P. X. K., Y. Fan, and J. D. Kalbfleisch. 2005. Maximization by parts in likelihood inference. J. Am. Stat. Assoc., 100(472), 1145–1167.
Todorov, V., N. Neyko, and P. Neytchev. 1994. Stability of high breakdown point robust PCA. Short Communications, COMPSTAT’94, Heidelberg, Germany: Physica Verlag.
Zhang, H. 2004. The optimality of naive Bayes. FLAIRS 2004 Conference. http://www.cs.unb.ca/profs/hzhang/publications/FLAIRS04ZhangH.pdf
Author information
Authors and Affiliations
Corresponding author
Additional information
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/ujsp.
Rights and permissions
About this article
Cite this article
Sen, S., Diawara, N. & Iftekharuddin, K.M. Statistical pattern recognition using Gaussian copula. J Stat Theory Pract 9, 768–777 (2015). https://doi.org/10.1080/15598608.2015.1008607
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1080/15598608.2015.1008607