Statistical Approaches of Ensembles: a Review

Inferential statistics is used to make predictions or comparisons about larger group (a population) using information gathered about a small part of that population.

• Discrete data are whole numbers, and are usually a count of objects. (For instance, one study might count how many pets different families own; it wouldn’t make sense to have half a goldfish, would it?) • Measured data, in contrast to discrete data, are continuous, and thus may take on any real value. (For example, the amount of time a group of children spent watching TV would be measured data, since they could watch any number of hours, even though their watching habits will probably be some multiple of 30 minutes.) • Numerical data are numbers. • Categorical data have labels (i.e. words). (For example, a list of the products bought by different families at a grocery store would be categorical data, since it would go something like {milk, eggs, toilet paper}.) [1] In mathematical physics, especially as introduced into statistical mechanics and thermodynamics by J. Willard Gibbs in 1878, an ensemble (also statistical ensemble or thermodynamic ensemble) [2,3] is an idealization consisting of a large number of mental copies (sometimes infinitely many) of a system, considered all at once, each of which represents a possible state that the real system might be in. Random matrix theory is a maturing discipline with decades of research in multiple _elds now beginning to converge. Experience has shown that many exact formulas are available for certain matrices with real, complex, or quaternion entries. In random matrix jargon, these are the cases _ = 1; 2 and 4 respectively. The three types of matrix models found predominantly in "classical" Random Matrix Theory are the Hermite (or Gaussian), Laguerre (or Wishart), and Jacobi (or MANOVA) ensembles. In this talk I will describe these ensembles and present some recent results in the study of eigenvalue distributions of the Hermite and Laguerre types, which were obtained using methods developed in Numerical Linear Algebra. The study of random matrices emerged in the late 1920's (with the publishing of Wishart's most important work in 1928) and 1930's (with Hsu's work [7]), and they are a very quickly growing _eld of research, with communities like nuclear physics [4, 5, 6], multivariate statistics [10, 8, 9], algebraic and enumerative combinatorics [11, 12, 13]. Many of the matrix models have thus standard normal entries, which are independent up to a symmetry/positive de_niteness condition (since the spectrum, whether it corresponds to the energy levels of a Schr•odinger operator or to the components of a sample covariance matrix, is real). Many studies of integrals over the general _ ensembles focused on the connection with Jack polynomials; of these we note the ones inspired by Selberg's work, like

2

We focus on the implications of this discovery to the point process limits of the spectral edge in the general _-ensembles. The distributional limits of the largest eigenvalues in G(O=U=S)E comprise some of the most celebrated results in random matrix theory due to their surprising importance in physics, combinatorics, multivariate statistics, engineering, and applied probability: [17-22] mark a few highlights.

The Gaussian ensembles were introduced by physicist Eugene Wigner in the 1950's. Though he started with a simpler model for a random matrix (entries from the uniform distribution on 1g, [23]) The general b ensembles appear to be connected to a broad spectrum of mathematics and physics, among which we list lattice gas theory, quantum mechanics, and Selberg-type integrals. Also, the b ensembles are connected to the theory of Jack polynomials ~with the correspondence a5 2/b where a is the Jack parameter!, which are currently objects of intensive research [ 24-26] Jack (1969-1970) originally defined the polynomials that eventually became associated with his name while attempting to evaluate an integral connected with the noncentral Wishart distribution [27,28]. Jack noted that the case were the Schur polynomials, and conjectured that were the zonal polynomials. The question of finding a combinatorial interpretation for the polynomials was raised by Foulkes [29], and subsequently answered by Knop and Sahi [30]. Later authors then generalized many known properties of the Schur and zonal polynomials to Jack polynomials [31,32].

We have reviewed the statistical approaches being implemented in ensembles for further work to investigate the standard output.

1. Keone Hon, An Introduction to Statistics, http://www.artofproblemsolving.com/LaTeX/Examples/statistics_firstfive.pdf 2. Kittel, Charles; Herbert Kroemer (1980). Thermal Physics, Second Edition. San Francisco: W.H. Freeman and Company. pp. 31 ff. ISBN 0-7167-1088-9. 5. 4. Freeman J. Dyson. The threefold way. Algebraic structures of symmetry groups and ensembles in Quantum Mechanics. J. Math. Phys., 3:1199{1215, 1963. 5. Michel Gaudin. Sur la loi limite de l'_espacement des valeurs propres d'une matrice al_eatoire. Nucl. Phys., 25:447{458, 1961. 6. Eugene P. Wigner. Characteristic vectors of bordered matrices with in_nite dimensions. Ann. of Math., 62:548{564, 1955. 7. P.L. Hsu. On the distribution of the roots of certain determinantal questions. Ann. Eugenics, 9:250{258, 1939. 8. Alan T. James. Distributions of matrix variates and latent roots derived from normal samples. Ann. Math. Stat., 35:475{501, 1964. 9. Robb J. Muirhead. Aspects of Multivariate Statistical Theory. John Wiley & Sons, New York, 1982. 10. A.G. Constantine. Some noncentral distribution problems in multivariate analysis. Ann. Math. Statist., 34:1270{1285, 1963. 11. Persi Diaconis. Group Representations in Probability and Statistics. Institute of Mathematical Statistics, Hayward, California, 1998. 12. I. Goulden and David M. Jackson. Maps in locally orientable surfaces and integrals over real symmetric matrices. Canadian J. Math., 49:865{882, 1997. 13. Philip J. Hanlon, Richard P. Stanley, and John R. Stembridge. Some combinatorial aspects of the spectra of normally distributed random matrices. Contemp. Math., 138:151{174, 1992. 14. K. Aomoto. Jacobi polynomials associated with Selberg integrals. SIAM J. Math. Anal., 18:545{549, 1987. 15. Joichi Kaneko. Selberg integrals and hypergeometric functions associated with Jack polynomials. SIAM J. Math. Anal., 24:1086{1110, 1993. 16. K. Kadell. The Selberg-Jack polynomials. Advances in Mathematics, 130:33{102, 1997. 17. Baik, J., Deift, P., Johansson, K. (1999) On the distribution of the length of the longest increasing

18. Baryshnikov, Y. (2001) GUEs and queues. Probab. Theory Rel. Fields. 119, no. 2, 256-274. 19. Ferrari, P. L., Spohn, H. (2006) Scaling limit for the space-time covariance of the stationary totally asymmetric simple exclusion process. Comm. Math. Phys. 265, no. 1, 1-44. 20. Johansson, K. (2000) Shape uctuations and random matrices. Comm. Math. Phys. 209, no. 2, 437-476. 21. Johnstone, I. M. (2001) On the distribution of the largest eigenvalue in principal components analysis. Ann. Statist. 29, no. 2, 295-327. 22. Prahofer, M., Spohn, H. ¨ (2002) Scale invariance of the PNG droplet and the Airy process. J. Statist. Phys. 108, no. 5-6, 1071-1106 23. Eugene P. Wigner. Characteristic vectors of bordered matrices with in_nite dimensions. Ann. of Math., 62:548{564, 1955. 24. Stanley, R. P., ‘‘Some combinatorial properties of Jack symmetric functions,’’ Adv. Math. 77, 76–115 ~1989!. 25. Macdonald, I., Symmetric Functions and Hall Polynomials ~Oxford University Press, New York, 1995!. 26. Okounkov, A. and Olshanski, G., ‘‘Shifted Jack polynomials, binomial formula, and applications,’’ Mathematical Research Letters 4, 69–78 ~1997!. 27. James, A. T. "The Distribution of the Latent Roots of the Covariance Matrix." Ann. Math. Stat. 31, 151-158, 1960.. 28. Hua, L. K. Harmonic Analysis of Functions of Several Complex Variables in the Classical Domains. Providence, RI: Amer. Math. Soc., 1963. 29. Foulkes, H. O. "A Survey of Some Combinatorial Aspects of Symmetric Functions." In Permutations. Paris: Gauthier-Villars, 1974. 30. Knop, F. and Sahi, S. "A Recursion and a Combinatorial Formula for the Jack Polynomials." Invent. Math. 128, 9-22, 1997. 31. Stanley, R. P. "Some Combinatorial Properties of Jack Symmetric Functions." Adv. in Math. 77, 76-115, 1989. University Press, pp. 383 and 387, 1995.