Abstract
While in theory many processes should have normal distribution or at least show asymptotic normality, in practice nonlinear systems often exhibit fat-tail distributions. Non-normality is hard to model, but also hard to detect (model selection). The problem is exacerbated by the varying complexity of the models, i.e. their propensity to overfit.
The Minimum Description Length principle applies Shannon’s information theory in statistical enquiry to balance between goodness of fit and model complexity. More specifically, the Normalized Maximum Likelihood (NML) model, stochastic distribution complexity are discussed.
Prior research has shown the distribution complexity for spherical distributions (uncorrelated identically distributed samples) in closed form.
The purpose of this paper is to extend the MDL framework to cover the independent samples case. A general and optimized numerical method for the calculation of the distribution complexity and the stochastic complexity is presented, with results shown for the Student-T distribution.
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
Kass, R., Raftery, A.: Bayes factors. Journal of the American Statistical Association 90(430), 773–795 (1995)
Nonchev, B.: Minimum Description Length Principle and Distribution Complexity of Spherical Distributions. In: Proceedings of the 18th European Young Statisticians Meeting (2013)
Kolmogorov, A.N.: On Tables of Random Numbers. Sankhya: The Indian Journal of Statistics, Series A 25, 369–376 (1963)
GrĂĽnwald, P.: The Minimum Description Length Principle. MIT Press (2007)
Rissanen, J.: Information and Complexity in Statistical Modeling (Information Science and Statistics). Springer (January 2007)
Shtarkov, Y.: Universal Sequential Coding of Single Messages. Problems of Information Transmission 23, 175–186 (1987)
Nonchev, B.: Minimum Description Length Principle in Discriminating Marginal Distributions. Pliska Studia Mathematica Bulgarica 22(125), 101–114 (2013)
Rissanen, J.: MDL Denoising. IEEE Transactions on Information Theory 46(7), 2537–2543 (2000)
Stine, R., Foster, D.: The Competitive Complexity Ratio. In: Proceedings of the 2001 Conference on Information Sciences and Systems. WP8, pp. 1–6 (2001)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Nonchev, B. (2014). Minimum Description Length Principle for Fat-Tailed Distributions. In: Mladenov, V.M., Ivanov, P.C. (eds) Nonlinear Dynamics of Electronic Systems. NDES 2014. Communications in Computer and Information Science, vol 438. Springer, Cham. https://doi.org/10.1007/978-3-319-08672-9_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-08672-9_10
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-08671-2
Online ISBN: 978-3-319-08672-9
eBook Packages: Computer ScienceComputer Science (R0)