In this paper an asset price model described by hidden Markov process dS(t)=μ(t)S(t)dt+σS(t)dW(t ) is considered, where W is a standard Brownian motion and σ is an unknown constant. The mean return{μ(t),0≤t≤T} is a stochastic process not necessarily adapted to the filtration generated by the process {S(t),0≤t≤T} and it contains  some unknown parameters to be estimated from a continuous time observation of S(t). Statistical estimators of the parameters σ and the parameters in μ based on Kalman filtering are  proposed and some numerical simulations are  performed  for the proposed estimators.