Bayesian shape analysis of the complex Bingham distribution

Abstract Previous shape analysis of landmark data arising from a complex Bingham distribution involved frequentist statistical estimation, and Bayesian MAP estimation. In this paper, we develop a full Bayesian analysis for planar landmark data arising from a complex Bingham. A novel MCMC method to sample the eigenvalues and eigenvectors of the complex-valued Hermitian parameter matrix is designed. Additionally, a new conjugate prior for the complex Bingham distribution is developed. Illustrative analysis using simulated and actual landmark data demonstrate the methods.

[1]  J. Kent.  The Complex Bingham Distribution and Shape Analysis , 1994 .

[2]  L. Scharf,et al.  Statistical Signal Processing of Complex-Valued Data: Notation , 2010 .

[3]  K. Mardia,et al.  The statistical analysis of shape data , 1989 .

[4]  K. Mardia,et al.  Statistical Shape Analysis , 1998 .

[5]  John Geweke,et al.  Evaluating the accuracy of sampling-based approaches to the calculation of posterior moments , 1991 .

[6]  Fred L. Bookstein,et al.  Morphometric Tools for Landmark Data. , 1998 .

[7]  Stephen G. Walker,et al.  Sampling from compositional and directional distributions , 2006, Stat. Comput..

[8]  F L Bookstein,et al.  Biometrics, biomathematics and the morphometric synthesis. , 1996, Bulletin of mathematical biology.

[9]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo conver-gence diagnostics: a comparative review , 1996 .

[10]  D.P. Mandic,et al.  Why a Complex Valued Solution for a Real Domain Problem , 2007, 2007 IEEE Workshop on Machine Learning for Signal Processing.

[11]  K. Mardia,et al.  The complex Watson distribution and shape analysis , 1999 .

[12]  D. Kendall.  The diffusion of shape , 1977, Advances in Applied Probability.

[13]  John T. Kent,et al.  Simulation for the complex Bingham distribution , 2004, Stat. Comput..

[14]  D. Kendall.  SHAPE MANIFOLDS, PROCRUSTEAN METRICS, AND COMPLEX PROJECTIVE SPACES , 1984 .

[15]  K. Mardia,et al.  The complex Bingham quartic distribution and shape analysis , 2006 .

[16]  Kazuyuki Aihara,et al.  Complex-valued forecasting of wind profile , 2006 .

[17]  Athanasios C. Micheas,et al.  Complex elliptical distributions with application to shape analysis , 2006 .

[18]  Peter D. Hoff,et al.  Simulation of the Matrix Bingham–von Mises–Fisher Distribution, With Applications to Multivariate and Relational Data , 2007, 0712.4166.

[19]  F. Bookstein.  Size and Shape Spaces for Landmark Data in Two Dimensions , 1986 .

[20]  J. Kent.  Data analysis for shapes and images , 1997 .

[21]  A. Wood,et al.  Empirical likelihood methods for two-dimensional shape analysis , 2010 .

[22]  Wael El-Deredy,et al.  Exploring event‐related brain dynamics with tests on complex valued time–frequency representations , 2008, Statistics in medicine.

[23]  Kazuyuki Aihara,et al.  Complex-valued prediction of wind profile using augmented complex statistics , 2009 .

[24]  Wilfrid S. Kendall,et al.  Alignments in two-dimensional random sets of points , 1980, Advances in Applied Probability.

[25]  C. Goodall.  Procrustes methods in the statistical analysis of shape , 1991 .