Optimal detection of a change-set in a spatial Poisson process (Q968778): Difference between revisions

The paper deals with a change-set problem for a spatial (\(2\)-dimensional) Poisson process. The problem has been inspired by results of \textit{T. Herberts} and \textit{U. Jensen} [Optimal detection of a change-point in a Poisson process for different observation schemes. Scand. J. Stat. 31, No. 3, 347--366 (2004; Zbl 1062.62154)], \textit{U. Jensen} and \textit{G.-H. Hsu} [Optimal stopping by means of point process observations with applications in reliability. Math. Oper. Res. 18, No.3, 645--657 (1993; Zbl 0778.60031)], \textit{G. Peskir} and \textit{A.N. Shiryaev} [Solving the Poisson disorder problem. K. Sandmann et al. (eds.), Advances in finance and stochastics. Essays in honour of D. Sondermann. Berlin: Springer, 295--312 (2002; Zbl 1009.60033)], and \textit{E. Bayraktar, S. Dayanik} and \textit{I. Karatzas}, [The standard Poisson disorder problem revisited. Stochastic Processes Appl. 115, No. 9, 1437--1450 (2005; Zbl 1070.62062)], where the \(1\)-dimensional optimal detection of disorder in the parameter of a Poisson process was studied. An exponential distribution of the change time was assumed. For the construction of solutions martingale techniques were applied. In a \(2\)--dimensional setting the authors formulated the optimal detection of the set by maximizing the expected value of a gain function. In the case that the unknown change-set is defined by a locally finite set of incomparable points they get a sufficient condition for optimal detection of the set using multi-parameter martingale techniques.