Modeling and Calibration for Some Stochastic Differential Models

In many scientific fields, the dynamics of system are often known, and main challenge is to estimate parameters that model behavior system. The question then arises whether one can use experimental measurements response derive parameters? This problem has been addressed in papers focus mainly on data from a deterministic model, but few efforts have made stochastic instead. this paper, we address using following procedure: first, build probabilistic differential models natural extension commonly used models. Then, by solving nonlinear regression problem. Since solutions not differentiable, well-known Nelder–Mead algorithm. Our numerical results show fitting procedure able obtain good estimates requiring only sample data.