Artigo Acesso aberto Revisado por pares

Deterministic version of Wigner's semicircle law for the distribution of matrix eigenvalues

1976; Elsevier BV; Volume: 13; Issue: 3 Linguagem: Inglês

10.1016/0024-3795(76)90095-1

ISSN

1873-1856

Autores

Ludwig Arnold,

Tópico(s)

Advanced Algebra and Geometry

Resumo

It is proved that Wigner's semicircle law for the distribution of eigenvalues of random matrices, which is important in the statistical theory of energy levels of heavy nuclei, possesses the following completely deterministic version. Let An=(aij), 1⩽i, ⩽n, be the nth section of an infinite Hermitian matrix, {λ(n)}1⩽k⩽n its eigenvalues, and {uk(n)}1⩽k⩽n the corresponding (orthonormalized column) eigenvectors. Let v∗n=(an1,an2,⋯,an,n−1), put Xn(t)=[n(n-1)]-12∑k=1[(n-1)t]|vn∗uf(n-1)|2,0⩽t⩽1 (bookeeping function for the length of the projections of the new row v∗n of An onto the eigenvectors of the preceding matrix An−1), and let finally Fn(x)=n-1(number of λk(n)⩽xn,1⩽k⩽n) (empirical distribution function of the eigenvalues of Ann. Suppose (i) limnannn=0, (ii) limnXn(t)=Ct(0<C<∞,0⩽t⩽1). Then Fn⇒W(·,C)(n→∞),where W is absolutely continuous with (semicircle) densityw(x,C)=(2Cπ)-1(4C-x212for|x|⩽2C0for|x|⩽2C

Referência(s)