This work presents a comparison of effectiveness of form parametric identification methods for linear dynamical systems operating in a stochastic environment. Three of these methods are correlation ones, and the other is derived from the least squares method. First of all, the system identification problem was introduced, and then, the structural aspects related to it were discussed. Later on, it was presented a review ot the above mentioned methods as well as a comparison between them. Such a comparison was concerned both with structural and computational aspects. This last stage uses simulated and real data.