No label defined (Q4940247)

The author considers the equation \[ y=\int_0^1 x(t) dw(t), \tag{1} \] where \(\{ w(t);t\in[0,1]\}\) is a Wiener process, \(x(t)\) is an unknown nonanticipative process, \(y\) is a random element which does not have finite second moment. The author proves that under some conditions there exists a unique stochastically smooth solution for equation (1). For example, the following assertion is proved: Let \(h\in L_2([0,1])\) and let a random variable \(\alpha\) be such that the transformation \(G(x)=x+i(h)\), \(x\in C_0([0,1])\), transforms \(\alpha\) into the twice stochastically differentiable in certain sense random variable. Then \(\alpha\) admits a unique representation \(\alpha=c+\int_0^1 x(t) dw(t)\), where the nonanticipative process \(\{ w(t); t\in[0,1]\}\) is twice stochastically differentiable.