EVALUATION OF THE EFFICIENCY OF APPLYING THE KALMAN FILTER TO DYNAMIC STOCHASTIC GENERAL EQUILIBRIUM MODELS

Mitina Olga Alekseevna, Gushel Vasiliy Olegovich

The article is devoted to the study of dynamic stochastic general equilibrium models and the evaluation of the effectiveness of applying the Kalman filter to them. The purpose of the study is to assess the applicability of such a tool as the Kalman filter for working with dynamic stochastic general equilibrium models in terms of filtering data and predicting the future state of the system based on the current one. The main method presented in the study is to compare the data obtained during the simulation of the model's behavior with the data obtained using the Kalman filter. In the study, this tool is used to clear data from distortions and predict the future state of the system. To determine the effectiveness of the Kalman filter, the analysis of various indicators is used in the work: linear Pearson correlation coefficient, mean and median values, standard deviation, standard and mean absolute error. In addition, a comparison with other filtering methods was made for a more objective assessment. The result of the study is the conclusion about the possibility of effective use of the Kalman filter when working with dynamic stochastic general equilibrium models, based on the analysis of the metrics under consideration and graphical representations of the filter.

Keywords:dynamic stochastic general equilibrium models, data filtering, Kalman filter, time series, data processing