scientific article; zbMATH DE number 3954782

From MaRDI portal
Publication:3724203





[For the entire collection see Zbl 0585.00014.] We suppose the evolution of the system described by a stochastic differential equation: (*) \(dx_ t=f(t,x,u)dt+\sigma (t,x)dB_ t\), \(x(o)=x_ o\). A first player \(J_ 1\) chooses a stopping time S, while a second player \(J_ 2\) chooses a continuous strategy u. There is an associated cost which acts until S, more precisely \(J(S,u)=E_ u(\int^{T}_{o}e^{-\alpha S}c(s,u_ s)ds+Y_ T)\) where \(E_ u\) is the expectation with respect to \(P_ u\) obtained by the Cameron-Martin formula. The player \(J_ 1\) (resp. \(J_ 2)\) acting with S (resp. u) looks for maximize (resp. minimize) the cost. It is then proved under the assumptions: (i) c measurable positive bounded and (ii) Y \(F_ t\)-adapted, right-continuous, bounded; that the game s described has always a value i.e.: \(\inf_{u} \sup_{s}\) \(J(S,u)=\sup_{s} \inf_{u}\) J(S,u).











This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3724203)