Date: Wednesday, March 21, 2018
Location: B844 East Hall (4:00 PM to 5:00 PM)
Title: Convergence of implicit schemes for HamiltonJacobiBellman quasivariational inequalities
Abstract: In [Azimzadeh, P., and P. A. Forsyth. "Weakly chained matrices, policy iteration, and impulse control." SIAM J. Num. Anal. 54.3 (2016): 13411364], we outlined the theory and implementation of computational methods for implicit schemes for HamiltonJacobiBellman quasivariational inequalities. No convergence proofs were given therein. This work closes the gap by giving rigorous proofs of convergence. A point of difficulty in the analysis is that a standard application of the BarlesSouganidis framework (BSF) requires a stronger comparison principle than that which is available in the literature. By introducing a stronger notion of consistency than that which is posed in the BSF, we are able to prove convergence relying only on a wellknown comparison principle. Our results are robust in that we do not assume a specific form for the intervention operator.
Joint work with Erhan Bayraktar and George Labahn.
Files:
Speaker: Parsiad Azimzadeh
Institution: UM
Event Organizer:
