Homogenization of locally stationary diffusions with possibly degenerate diffusion matrix (Q985340)

The paper deals with homogenization of second order PDEs with locally stationary coefficients. Namely, it is considered the strong solution \(X^\varepsilon\) of the following stochastic differential equation \[ X^\varepsilon_t= x+{1\over\varepsilon} \int^t_0 b\Biggl({X^\varepsilon_r\over\varepsilon}, X^\varepsilon_r\Biggr)\,dr+ \int^t_0 c\Biggl({X^\varepsilon_r\over \varepsilon}, X^\varepsilon_r\Biggr)\,dr+ \int^t_0\sigma \Biggl({X^\varepsilon_r\over\varepsilon}, X^\varepsilon_r\Biggr)\, dB_r, \] where for each \(y\in\mathbb{R}^d\) coefficients \(b(\cdot,y)\), \(c(\cdot,y)\), \(\sigma(\cdot,y)\) are stationary random fields independent of the \(d\)-dimensional standard Brownian motion \(B\). Thus \(b\), \(c\) and \(\sigma\) take into account both microscopic and macroscopic evolution scales. It is proved that, under suitable assumptions, the process \(X^\varepsilon\) converges weakly as \(\varepsilon\to 0\) in \(C([0, T];\mathbb{R}^d)\) to the solution of a stochastic differential equation with deterministic coefficients. The paper follows and improves the author's earlier paper [Probab. Theor. Related Fields, 143, 545--568 (2009; Zbl 1163.60049)] by considering possibly degenerate diffusion matrices.