Abstract
The iterative fixed posteriori approximation (iterative FPA) has been empirically shown to be an efficient approach for the MAP factor estimate in the Non-Gaussian Factor Analysis (NFA) model. In this paper we further prove that it is exactly an EM algorithm for the MAP factor estimate problem. Thus its convergence can be guaranteed. We also empirically show that NFA has better generalization ability than Independent Factor Analysis (IFA) on data with small sample size.
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
Xu, L.: Byy harmony learning, independent state space and generalized apt financial analysis. IEEE Transaction on Neural Network 12 (2001) 822–849
Xu, L.: Bayesian kullback ying-yang dependence reduction theory. Neurocomputing 19 (1998) 223–257
Liu, Z.Y., Xu, L.: A comparative study on three map factor estimate approaches in non-gaussian factor analysis model. International Conference on Intelligent Data Engineering and Automated Learning (IDEAL’02) (Manchester, UK, 2002)
Wu, C.: On the convergence properties of the em algorithm. The Annals of Statistics 11 (1983) 95–103
Ma, J., Xu, L., Jordan, M.: Asymptotic convergence rate of the em algorithm for gaussian mixtures. Neural Computation 12 (2000) 2881–2907
Xu, L., Jordan, M.: On convergence properties of the em algorithm for gaussian mixture. Neural Computation 8 (1996) 129–151
Dempster, A., Laird, N., Rubin, D.: Maximum-likelihood from incomplete data via the em algorithm. J. Royal Statist. Soc. Ser. B. 39 (1977) 1–38
Attias, H.: Independent factor analysis. Neural Computation 11 (1999) 803–851
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Liu, Z., Xu, L. (2002). On Convergence of an Iterative Factor Estimate Algorithm for the NFA Model. In: Dorronsoro, J.R. (eds) Artificial Neural Networks — ICANN 2002. ICANN 2002. Lecture Notes in Computer Science, vol 2415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46084-5_166
Download citation
DOI: https://doi.org/10.1007/3-540-46084-5_166
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44074-1
Online ISBN: 978-3-540-46084-8
eBook Packages: Springer Book Archive