Abstract
In high-dimensional directional statistics one of the most basic probability distributions is the von Mises-Fisher (vMF) distribution. Maximum likelihood estimation for the vMF distribution turns out to be surprisingly hard because of a difficult transcendental equation that needs to be solved for computing the concentration parameter κ. This paper is a followup to the recent paper of Tanabe et al. (Comput Stat 22(1):145–157, 2007), who exploited inequalities about Bessel function ratios to obtain an interval in which the parameter estimate for κ should lie; their observation lends theoretical validity to the heuristic approximation of Banerjee et al. (JMLR 6:1345–1382, 2005). Tanabe et al. (Comput Stat 22(1):145–157, 2007) also presented a fixed-point algorithm for computing improved approximations for κ. However, their approximations require (potentially significant) additional computation, and in this short paper we show that given the same amount of computation as their method, one can achieve more accurate approximations using a truncated Newton method. A more interesting contribution of this paper is a simple algorithm for computing I s (x): the modified Bessel function of the first kind. Surprisingly, our naïve implementation turns out to be several orders of magnitude faster for large arguments common to high-dimensional data, than the standard implementations in well-established software such as Mathematica ©, Maple ©, and Gp/Pari.
References
Abramowitz, M, Stegun, IA (eds) (1974) Handbook of mathematical functions, with formulas, graphs, and mathematical tables. Dover, New York ISBN 0486612724
Amos DE (1974) Computation of modified Bessel functions and their ratios. Math Comput 28(125): 235–251
Banerjee A, Dhillon IS, Ghosh J, Sra S (2005) Clustering on the unit hypersphere using von Mises-Fisher distributions. JMLR 6: 1345–1382
Dhillon IS, Sra S (2003) Modeling data using directional distributions. Technical Report TR-03-06, Computer Sciences, The University of Texas at Austin
Gourdon X, Sebah P (2002) Convergence acceleration of series. http://numbers.computation.free.fr/Constants/constants.html
Mardia KV, Jupp P (2000) Directional statistics, second edition. Wiley, London
Maxima (2008) http://maxima.sourceforge.net. Computer Algebra System version 5.16.3
MPFR (2008) http://www.mpfr.org. Multi-precision floating-point library version 2.3.1
PARI/GP (2008) version \({{\tt 2.3.4}}\). The PARI Group, Bordeaux. available from http://pari.math.u-bordeaux.fr/
Sra S (2007) Matrix nearness problems in data mining. PhD thesis, University of Texas at Austin
Tanabe A, Fukumizu K, Oba S, Takenouchi T, Ishii S (2007) Parameter estimation for von Mises-Fisher distributions. Comput Stat 22(1): 145–157
Author information
Authors and Affiliations
Corresponding author
Electronic Supplementary Material
Rights and permissions
About this article
Cite this article
Sra, S. A short note on parameter approximation for von Mises-Fisher distributions: and a fast implementation of I s (x). Comput Stat 27, 177–190 (2012). https://doi.org/10.1007/s00180-011-0232-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-011-0232-x