Numerical Simulation on Dynamic and Stochastic Approach of General Equilibrium

• Grant number: 15/07415-7
• Support Opportunities: Scholarships abroad - Research
• Start date: September 01, 2015
• End date: February 29, 2016
• Field of knowledge: Applied Social Sciences - Economics - Quantitative Methods Applied to Economics
• Principal Investigator: Ricardo Luis Chaves Feijó
• Grantee: Ricardo Luis Chaves Feijó
• Host Investigator: Wolfgang Hardle
• Host Institution: Faculdade de Economia, Administração e Contabilidade de Ribeirão Preto (FEARP). Universidade de São Paulo (USP). Ribeirão Preto , SP, Brazil
• Institution abroad: Humboldt University, Germany

Abstract

The future research aims to enhance the numerical simulation techniques of dynamic general equilibrium in order to adapt the model to a more complex, realistic and practical scenario. With new computational statistical methods we intended to achieve a numerical solution to improve the meet of the subsidiaries conditions of the problem. It is intended to better identify the nature of the singularity numerically found to measure the probabilities involved in the point selection process, so ensures the uniqueness found to be statistically significant in relation to stochastic border. In the model extension, it is intended to work with models in which the agents have different utility functions. We intend to improve, with the help of German professors, our simulation techniques in order to incorporate more advanced models. It is believed also to incorporate in the technique the computational approach for models with infinite life agents, thus significantly improving the quality of the result of the equilibrium of the stochastic simulations. Another important step in the extension of technique, is to learn to model, for the purpose of numerical simulations, situations in which does not impose any restriction on the exogenous level of indebtedness of the agents. One way to avoid this kind of artificiality is to incorporate the use of collateral in transactions involving the purchase and sale of financial assets. Therefore, we intend to develop more sophisticated models for collateralised credit, and improved our numerical simulation technique to incorporate such models. We also intend to extend the simulation model to address the phenomenon of default by agents.