Showing 1–3 of 3 results for author: Sørensen, T M

Stability of an upwind Petrov Galerkin discretization of convection diffusion equations

Authors: Snorre H. Christiansen, Tore G. Halvorsen, Torquil M. Sørensen

Abstract: We study a numerical method for convection diffusion equations, in the regime of small viscosity. It can be described as an exponentially fitted conforming Petrov-Galerkin method. We identify norms for which we have both continuity and an inf-sup condition, which are uniform in mesh-width and viscosity, up to a logarithm, as long as the viscosity is smaller than the mesh-width or the crosswind dif… ▽ More We study a numerical method for convection diffusion equations, in the regime of small viscosity. It can be described as an exponentially fitted conforming Petrov-Galerkin method. We identify norms for which we have both continuity and an inf-sup condition, which are uniform in mesh-width and viscosity, up to a logarithm, as long as the viscosity is smaller than the mesh-width or the crosswind diffusion is smaller than the streamline diffusion. The analysis allows for the formation of a boundary layer. △ Less

Submitted 22 February, 2016; v1 submitted 2 June, 2014; originally announced June 2014.

Comments: v1: 18 pages. 2 figures. v2: 22 pages. Numerous details added and completely rewritten final proof. 8 pages appendix with old proof

MSC Class: 65N30

Comment on "The Stochastic Nonlinear Schrödinger Equation in $H^{1}$"

Authors: Torquil Macdonald Sørensen

Abstract: The paper "The Stochastic Nonlinear Schrödinger Equation in $H^{1}$" \cite{debouard2003} gives an existence proof for a stochastic nonlinear Schrödinger equation with multiplicative noise. We point out two mistakes that draw the validity of the proof into question. The paper "The Stochastic Nonlinear Schrödinger Equation in $H^{1}$" \cite{debouard2003} gives an existence proof for a stochastic nonlinear Schrödinger equation with multiplicative noise. We point out two mistakes that draw the validity of the proof into question. △ Less

Submitted 25 February, 2012; originally announced February 2012.

Comments: 2 pages

MSC Class: 35R60 (Primary) 35Q41; 35G20 (Secondary)

Levy process simulation by stochastic step functions

Authors: Torquil Macdonald Sørensen, Fred Espen Benth

Abstract: We study a Monte Carlo algorithm for simulation of probability distributions based on stochastic step functions, and compare to the traditional Metropolis/Hastings method. Unlike the latter, the step function algorithm can produce an uncorrelated Markov chain. We apply this method to the simulation of Levy processes, for which simulation of uncorrelated jumps are essential. We perform numerical… ▽ More We study a Monte Carlo algorithm for simulation of probability distributions based on stochastic step functions, and compare to the traditional Metropolis/Hastings method. Unlike the latter, the step function algorithm can produce an uncorrelated Markov chain. We apply this method to the simulation of Levy processes, for which simulation of uncorrelated jumps are essential. We perform numerical tests consisting of simulation from probability distributions, as well as simulation of Levy process paths. The Levy processes include a jump-diffusion with a Gaussian Levy measure, as well as jump-diffusion approximations of the infinite activity NIG and CGMY processes. To increase efficiency of the step function method, and to decrease correlations in the Metropolis/Hastings method, we introduce adaptive hybrid algorithms which employ uncorrelated draws from an adaptive discrete distribution defined on a space of subdivisions of the Levy measure space. The nonzero correlations in Metropolis/Hastings simulations result in heavy tails for the Levy process distribution at any fixed time. This problem is eliminated in the step function approach. In each case of the Gaussian, NIG and CGMY processes, we compare the distribution at t=1 with exact results and note the superiority of the step function approach. △ Less

Submitted 11 October, 2011; originally announced October 2011.

Comments: 20 pages, 18 figures

MSC Class: 60G51 (Primary) 65C05; 60G17; 62M10; 60J22; 60J75 (Secondary) ACM Class: G.3

Journal ref: SIAM Journal of Scientific Computing, Volume 35, Issue 5, Pages 2207-2224, 2013