3-D converted-wave data were acquired using digital MEMS (micro-electromechanical system) three component (3C) sensors in the alternating sand and shale sequence in the overburden of the Shengli Ken-71 area. This gives rise to serious non-hyperbolic moveout effects in the converted-wave data due to both the asymmetrical ray path and anisotropic effects. Conventional velocity analysis and moveout correction based on isotropic methods do not flatten reflections events. Here, we use a four-parameter theory to evaluate these effects and process the data. These four parameters include the PS converted wave stacking velocity (VC2), the vertical velocity ratio (y0), the effective velocity ratio (yeff), and the anisotropy parameter (xeff). The method utilizes the moveout information at different offsets to estimate the different parameters and ensures that the events are properly aligned for stacking. As a result, this four-parameter theory leads to an improvement in imaging quality and correlation between the P-waves and converted-waves.