Parameter estimation for multiscale diffusions (Q2370014)

This manuscript is devoted to the study of parameter estimation for time-series possessing two, widely separated, characteristic time scales. It is proved that if the data is sampled too finely then the parameter fit will fail, in that the correct parameters in the homogenized model are not identified. It is also proved that if the data is sub-sampled at an appropriate rate then it is possible to estimate the coefficients of the homogenized model correctly. Numerical and analytical proves are presented.