The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
The search session has expired. Please query the service again.
Consider a one dimensional nonlinear reaction-diffusion equation
(KPP equation) with non-homogeneous second order term, discontinuous
initial condition and small parameter. For points ahead
of the Freidlin-KPP front, the solution tends to 0 and we obtain
sharp asymptotics (i.e. non logarithmic). Our study follows the
work of Ben Arous and Rouault who solved this problem in the
homogeneous case. Our proof is probabilistic, and is based on
the Feynman-Kac formula and the large deviation principle...
We start with a general time-homogeneous scalar diffusion whose state space is an interval I ⊆ ℝ. If it is started at x ∈ I, then we consider the problem of imposing upper and/or lower boundary conditions at two points a,b ∈ I, where a < x < b. Using a simple integral identity, we derive general expressions for the Laplace transform of the transition density of the process, if killing or reflecting boundaries are specified. We also obtain a number of useful expressions for the Laplace transforms...
Currently displaying 1 –
6 of
6