EN
The return on sale (ROS) is one of the most important indexes used to estimate the effectiveness of running an enterprise. This index is usually estimated “at point”, having information about financial results and sales profits of an enterprise. By that it means, it does not include the developing path of an enterprise, which means that the results of work in period “t” do not affect the results of work in earlier periods. In analysis below, we suggest to estimate ROS by using the regression model with various variables. It can help overcome the inconvenience stated above. ROS can be estimated by using the following pattern: ROS_{t} = 1 - [( a_{0t}/PS_{t}) ) + a_{1t}] The results show, that this model has a lot of cognitive values because it helps realize the calculus of return on sale. This model can easily be used to make ROS prognosis.