Comparison theorem for strong solution of two-parameter Poisson type stochastic differential equation (Q917162)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Comparison theorem for strong solution of two-parameter Poisson type stochastic differential equation |
scientific article |
Statements
Comparison theorem for strong solution of two-parameter Poisson type stochastic differential equation (English)
0 references
1990
0 references
For a fixed point \(t_ 0=(t'_ 0,t^ 2_ 0)\in {\mathbb{R}}^ 2_+\), consider two-parameter Poisson type SDEs as follows: \[ (1)\quad dX_ t=\alpha_ i(X_ t)d\Lambda_ t+\beta (X_ t)dN_ t,\quad X_ t=Z,\quad t\in \{0\}\times [0,t^ 2_ 0]\cup [0,t'_ 0]\times \{0\},\quad i=1,2, \] \(t\in R_{t_ 0}=\{t'\), \(t'\leq t_ 0\}\), \(N(t)=Y(t)-\Lambda (R_ t)\), where \(\Lambda\) is a nonnegative, additive set function on \({\mathbb{R}}^ 2_+\), and Y(t) is a two-parameter Poisson random measure with parameter \(\lambda\) ([0,t]). It is easy to see that \(\{\) N(t), \(t\in R_{t_ 0}\}\) is a two-parameter strong martingale with respect to the natural \(\sigma\)-field \(\{\) \({\mathcal F}_ t\}\). Suppose \(X^ 1(t)\), \(X^ 2(t)\) are solutions of the SDE, \[ X^ i(t)=\int_{R_ t}\alpha_ i(X^ i(s))d\Lambda_ s+\int_{R_ t}\beta (X^ i(s))dN(s),\quad i=1,2. \] If (i) there exists a strict increasing, lower convex function \(\rho\), \(\rho (0)=0\), such that \(| \beta (x)-\beta (y)| \leq \rho (| x-y|)\), \[ \int^{\delta}_{0+}(u+\rho (u))^{-1}du=+\infty,\quad \forall \delta >0, \] (ii) there are two real-valued, continuous, nondecreasing functions \(b_ 1(x)\), \(b_ 2(x)\) satisfying \(\alpha_ 1(x)\leq b_ 1(x)\), \(\alpha_ 2(x)\geq b_ 2(x)\), and \(b_ 1(x)<b_ 2(x)\forall x\in R'\), then \[ P(X^ 1(t)\leq X^ 2(t),\text{ for all } t\in R_{t_ 0})=1. \]
0 references
comparison theorem for SDE
0 references
two-parameter Poisson random measure
0 references
two- parameter strong martingale
0 references