next up previous
Next: Expressions in Terms of Up: Simple Explicit Formulas for Previous: Wronski's Construction of Green

Main Results and Relation to Gelfand-Yaglom's Initial-Value Problem

Excluding at first zero modes, we evaluate formula (17) for ratios of functional determinants. The temporal integral on the right-hand side can be performed efficiently following Ref. [10]. Here we present an even more direct method, by which we express the result in terms of solutions of Gelfand-Yaglom's initial-value problem for Dirichlet boundary conditions, and of a dual problem for periodic and antiperiodic boundary conditions.  
 
Dirichlet Case
The Gelfand-Yaglom initial-value problem consists in the search for a function tex2html_wrap_inline2259 solving the following equations:
 equation366
By differentiating these three equations with respect to the parameter g, we obtain for tex2html_wrap_inline2263 the inhomogeneous initial-value problem
 equation370
The unique solution of equations (41) can easily be expressed in terms of our arbitrary set of solutions tex2html_wrap_inline2185 and tex2html_wrap_inline2187 as follows
 equation377
thus leading to
 equation382
In terms of the same functions, the general solution of the inhomogeneous initial-value problem (42) can be seen to have the form
 equation388
Comparison with (34) shows that at the final point tex2html_wrap_inline2269
 equation399
which together with (44) implies the following simple relation for the Green function (34) with Dirichlet's boundary conditions:
 equation407
Inserting this into (17), we find for the ratio of functional determinants the simple formula
 equation415
The constant of integration is fixed by applying (48) to the trivial case g=0, where tex2html_wrap_inline2273 and the solution to the initial-value problem (41) is
 equation422
At g=0, the left-hand side of (50) is unity, determining tex2html_wrap_inline2277 and the final result for g=1:
 equation427
This compact formula was first derived by Gelfand and Yaglom [2] via a direct calculation of the determinant arising in a time-sliced path integrals [1].

 
Periodic and Antiperiodic Case
Our technique makes it straight-forward to derive an equally compact formula for periodic and antiperiodic boundary conditions. For this purpose we introduce another homogeneous initial-value problem whose boundary conditions are dual to Gelfand and Yaglom's in (41):
 equation443
In terms of the previous arbitrary set tex2html_wrap_inline2185 and tex2html_wrap_inline2187 of solutions of the homogeneous differential equation, the unique solution of (51) reads
 equation448
This can be combined with the time derivative of (43) at tex2html_wrap_inline2269 to yield
 equation454
By differentiating Eqs. (51) with respect to g, we obtain the following inhomogeneous initial-value problem for tex2html_wrap_inline2289:
 equation460
whose general solution reads in analogy to (45)
 equation468
where the dot denotes the time derivative with respect of the first argument of tex2html_wrap_inline2169. With the help of identities (26) and (27), the combination tex2html_wrap_inline2293 at tex2html_wrap_inline2269 can now be expressed in terms of the periodic and antiperiodic Green functions (40), in analogy to (46),
 equation485
Together with (53), this yields for the temporal integral on the right-hand sides of (16) and (17) the simple expression analogous to (47)
 eqnarray497
This is inserted into formula Eq. (17) yielding for periodic and antiperiodic boundary conditions
 equation505
where tex2html_wrap_inline2147. The constant of integration C is fixed in the way described after Eq. (17). We go to g=1 and set tex2html_wrap_inline2303. For the operator tex2html_wrap_inline2305, we can easily solve the Gelfand-Yaglom initial-value problem (41) as well as the dual one (51) by
 equation514
so that (58) determines C by
 equation523
Hence we obtain the final results for periodic boundary conditions
 eqnarray532
and for antiperiodic boundary conditions
 eqnarray544
The intermediate expressions in (50), (61), and (62) show that the ratios of functional determinants are ordinary determinants of two arbitrary independent solutions tex2html_wrap_inline2309 and tex2html_wrap_inline2311 of the homogeneous differential equation tex2html_wrap_inline2313. As such, the results are manifestly invariant under arbitrary linear transformations of these functions tex2html_wrap_inline2315.


next up previous
Next: Expressions in Terms of Up: Simple Explicit Formulas for Previous: Wronski's Construction of Green

Hagen Kleinert
Fri Jul 10 10:10:11 MET DST 1998