The main purpose of the present paper is to formulate the maximum likelihoo
d state estimation problem correctly for a continuous-time nonlinear stocha
stic dynamical system. By using the Onsager-Machlup functional, a modified
likelihood is introduced. The basic equation for the maximum likelihood sta
te estimate is derived with the aid of a dynamic programming approach. The
numerical procedure for realizing the recursive filtering is also proposed
with some numerical results.