Krein's trace theorem revisited
Author(s) -
Denis Potapov,
Fedor Sukochev,
Dmitriy Zanin
Publication year - 2014
Publication title -
journal of spectral theory
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.063
H-Index - 19
eISSN - 1664-0403
pISSN - 1664-039X
DOI - 10.4171/jst/75
Subject(s) - trace (psycholinguistics) , mathematical economics , mathematics , calculus (dental) , philosophy , linguistics , medicine , dentistry
M. G. Krein’s celebrated Trace Theorem” states that if A ,B are self-adjoint operators in a separable Hilbert space such that A − B is a trace class operator, then, for any function f of a real variable, whose derivative f ′ in distributional sense has Fourier transform belonging to L1(R) , the difference f(A)− f(B) is again a trace class operator and the formula Trace ( f(A)− f(B) ) = ∫ R f ′(s)ξ(s)ds holds where the function ξ ∈ L1(R) depends only on A and B and is uniquely determined by the above formula. The function ξ is called the spectral shift function of the pair A ,B and is an important ingredient in the perturbation theory of self-adjoint operators. The original proof of Krein uses complex analysis and is quite involved. We supply a new proof which does not use complex analysis. Our proof works also for σ-finite von Neumann algebrasM of type II and unbounded perturbations from the predual ofM. The exposition is based on joint work with D. Potapov and D. Zanin. E-mail address: f.sukochev@unsw.edu.au School of Mathematics & Statistics, University of NSW, Kensington NSW 2052 AUSTRALIA
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom