Estimation of Multiproduct Models in Economics on the Example of Production Sector of Russian Economy
The model of the real sector of the Russian economy is presented. It allows for the separate description of GDP and its components by expenditure both in constant and in current prices. Unlike standard macroeconomic models, the model proposed considers a set of Trader agents in addition to Producer agent. Traders are based on a set of CES-functions and allow to decompose the statistics available into a set of unobserved components. The Producer is based on a specific production function that performs well for Russian data and works with financial variables, such as credits and bank accounts. In contrary to the standard approach, the model is not linearized to get estimates of model parameters but is estimated directly using a set of nonlinear equations. The optimization is performed numerically and allows to get both series of unobserved model products and their prices and model parameters. The stability of the solution found is checked on simulated data.
KeywordsMacroeconomic modeling Nonlinear models Gross Domestic Product (GDP) Mathematical programming
- 2.Andreev, M., Pilnik, N., Pospelov, I.G., Vrzheshch, V.P., Masyutin, A.: Intertemporal three-product general equilibrium model of Russian economy. Int. J. Arts Sci. 6(1), 125–145 (2013)Google Scholar
- 5.Chatterjee, S., Cooper, R.: Entry and exit, product variety and the business cycle (No. w4562). National Bureau of Economic Research (1993)Google Scholar
- 7.Dixit, A.K., Stiglitz, J.E.: Monopolistic competition and optimum product diversity. Am. Econ. Rev. 67(3), 297–308 (1977)Google Scholar
- 12.Pilnik, N., Pospelov, I., Stankevich, I.: Multiproduct Model Decomposition of Components of Russian GDP. NRU Higher School of Economics. Series WP BRP “Economics/EC”. No. WP BRP 111/EC/2015 (2015)Google Scholar
- 13.Pilnik, N., Radionov, S.: On new approaches to the identification of blocks of general equilibrium models. Proc. MIPT 3(35), 151–161 (2017)Google Scholar