The paper presents a refined regression model of water level dynamics in the Siberian river Iya, which includes six natural factors on the right side (the number of days with precipitation in the Sayan Mountains, average day and night temperatures for the month, the amount of precipitation, snow depth, average atmospheric pressure for the month ) taking into account the delay, as well as a specially generated seasonal variable. The high adequacy of the model is indicated by the values of the criteria of multiple determination, Fisher, and the average relative error of approximation. The constructed model can be effectively used to solve a wide range of forecasting problems.
Keywords: regression model, river water level, lag time, seasonal variable, forecast