No label defined (Q4394491)

Physical systems in actual situations are so complicated that they usually need some approximation. In path integrals, the most promising one seems to be a semiclassical approximation. And there are some systems in which the WKB approximation gives the exact result such as a harmonic oscillator. Recently, a new possibility of exactness of the WKB approximation is discussed in connection with the theorem of Duistermaat-Heckman about \(SU(2)\) spin system or \(Cp^N\) system. But there are a number of unsatisfactory points and mysteries in the discussion: starting action being already semiclassical, subtlety of the use of continuum path integral under the coherent state representation, the so-called overspecification problem, etc. Motivated thus we study the exactness of the WKB approximation under the path integral formula in the case of \(SU(2)\) and \(SU(1,1)\) with the aid of generalized coherent states, and in this paper we concentrate the WKB exactness in our method. For the entire collection see [Zbl 0884.00037].