Bartolucci, Francesco (2011): An alternative to the BaumWelch recursions for hidden Markov models.

PDF
MPRA_paper_38778.pdf Download (271kB)  Preview 
Abstract
We develop a recursion for hidden Markov model of any order h, which allows us to obtain the posterior distribution of the latent state at every occasion, given the previous h states and the observed data. With respect to the wellknown BaumWelch recursions, the proposed recursion has the advantage of being more direct to use and, in particular, of not requiring dummy renormalizations to avoid numerical problems. We also show how this recursion may be expressed in matrix notation, so as to allow for an efficient implementation, and how it may be used to obtain the manifest distribution of the observed data and for parameter estimation within the ExpectationMaximization algorithm. The approach is illustrated by an application to nancial data which is focused on the study of the dynamics of the volatility level of logreturns.
Item Type:  MPRA Paper 

Original Title:  An alternative to the BaumWelch recursions for hidden Markov models 
Language:  English 
Keywords:  ExpectationMaximization algorithm, forwardbackward recursions, latent Markov model, stochastic volatility 
Subjects:  C  Mathematical and Quantitative Methods > C1  Econometric and Statistical Methods and Methodology: General > C13  Estimation: General C  Mathematical and Quantitative Methods > C2  Single Equation Models; Single Variables > C23  Models with Panel Data; Longitudinal Data; Spatial Time Series C  Mathematical and Quantitative Methods > C2  Single Equation Models; Single Variables > C22  TimeSeries Models; Dynamic Quantile Regressions; Dynamic Treatment Effect Models 
Item ID:  38778 
Depositing User:  Francesco Bartolucci 
Date Deposited:  13. May 2012 17:14 
Last Modified:  14. Feb 2013 16:31 
References:  Bartolucci, F. and Besag, J. (2002). A recursive algorithm for Markov random fields. Biometrika, 89:724730. Bartolucci, F., Farcomeni, A., and Pennoni, F. (2010). An overview of latent Markov models for longitudinal categorical data. Technical report available at http://arxiv.org/abs/1003.2804. Baum, L. E., Petrie, T., Soules, G., and Weiss, N. (1970). A maximization technique occurring in the statistical analysis of probabilistic functions of Markov chains. Annals of Mathematical Statistics, 41:164171. Dempster, A. P., Laird, N. M., and Rubin, D. B. (1977). Maximum likelihood from incomplete data via the EM algorithm (with discussion). Journal of the Royal Statistical Society, Series B, 39:138. Juang, B. and Rabiner, L. (1991). Hidden Markov models for speech recognition. Technometrics, 33:251272. Lystig, T. C. and Hughes, J. (2002). Exact computation of the observed information matrix for hidden Markov models. Journal of Computational and Graphical Statistics, 11:678689. Schwarz, G. (1978). Estimating the dimension of a model. Annals of Statistics, 6:461464. Scott, S. L. (2002). Bayesian methods for hidden Markov models: Recursive computing in the 21st century. Journal of the American Statistical Association, 97:337351. Taylor, S. J. (2005). Asset Price Dynamics, Volatility, and Prediction. Princeton University Press, Princeton. Welch, L. R. (2003). Hidden Markov models and the BaumWelch algorithm. IEEE Information Theory Society Newsletter, 53:113. Zucchini, W. and MacDonald, I. L. (2009). Hidden Markov and other Models for Time Series: An Introduction Using R. Chapman and Hall, CRC. 
URI:  http://mpra.ub.unimuenchen.de/id/eprint/38778 