摘要：AbstractThe paper is devoted to analysis of one-sector growth models and corresponding control problems on optimal distribution of investments. The paper considers a model with a linear production function, which takes into account the feasibility of structural changes in an economy. By introducing dummy variables one can statistically indicate a period when the model undergoes changes. This provides the possibility to switch the model in different modes for providing more accurate forecasts of economic development. For the optimal control problems, the qualitative analysis of the Hamiltonian systems is implemented and solutions are constructed for each model mode. Continuous gluing of the obtained trajectories is obtained as a solution of the optimal control problem with different model modes on the infinite time interval. Comparison of the resulting model trajectory with statistical data reveals that the simulated trends provide sufficiently accurate matching with the real data. Adaptation of model parameters to a new economic mode can be considered as alearning processfor the entire optimal control model. It makes the model more flexible with respect to the qualitative changes influencing forecasts of economic development.
关键词：Keywordsoptimal control problemeconomic growthHamiltonian trajectorieseconometric data analysisdummy variablesswitching model modes
