Computational Statistics

, Volume 27, Issue 1, pp 177–190 | Cite as

A short note on parameter approximation for von Mises-Fisher distributions: and a fast implementation of I s (x)

  • Suvrit SraEmail author
Short Note


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.


von Mises-Fisher distribution Maximum-likelihood Numerical approximation Modified Bessel function Bessel ratio 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Supplementary material

180_2011_232_MOESM1_ESM.h (0 kb)
ESM 1 (1 kb) (1 kb)
ESM 2 (2 kb)


  1. Abramowitz, M, Stegun, IA (eds) (1974) Handbook of mathematical functions, with formulas, graphs, and mathematical tables. Dover, New York ISBN 0486612724Google Scholar
  2. Amos DE (1974) Computation of modified Bessel functions and their ratios. Math Comput 28(125): 235–251MathSciNetGoogle Scholar
  3. Banerjee A, Dhillon IS, Ghosh J, Sra S (2005) Clustering on the unit hypersphere using von Mises-Fisher distributions. JMLR 6: 1345–1382MathSciNetzbMATHGoogle Scholar
  4. Dhillon IS, Sra S (2003) Modeling data using directional distributions. Technical Report TR-03-06, Computer Sciences, The University of Texas at AustinGoogle Scholar
  5. Gourdon X, Sebah P (2002) Convergence acceleration of series.
  6. Mardia KV, Jupp P (2000) Directional statistics, second edition. Wiley, LondonzbMATHGoogle Scholar
  7. Maxima (2008) Computer Algebra System version 5.16.3
  8. MPFR (2008) Multi-precision floating-point library version 2.3.1
  9. PARI/GP (2008) version \({{\tt 2.3.4}}\). The PARI Group, Bordeaux. available from
  10. Sra S (2007) Matrix nearness problems in data mining. PhD thesis, University of Texas at AustinGoogle Scholar
  11. Tanabe A, Fukumizu K, Oba S, Takenouchi T, Ishii S (2007) Parameter estimation for von Mises-Fisher distributions. Comput Stat 22(1): 145–157MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  1. 1.Max-Planck Institute (MPI) for biological CyberneticsTübingenGermany

Personalised recommendations