Correlações espaciais e temporais da dinâmica de distribuição de renda

• Autor(a) principal: LIMA, Ayron Andrey da Silva
• Data de Publicação: 2019
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UFRPE
• Texto Completo: http://www.tede2.ufrpe.br:8080/tede2/handle/tede2/8521
• topic: Distribuição de renda; Correlação espacial; Correlação temporal; Modelo estocástico; CIENCIAS EXATAS E DA TERRA::FISICA

In the last three decades much effort has been made by researchers and public agents to understand the role of the spatial distribution of income in socio-environmental development at various scales such as neighborhoods, cities, metropolitan regions, countries and even continents. For different time scales and income regimes, a large class of models, belonging to an area that is conventionally called economophysical, can be found in the literature, from those that incorporate spatial and topological aspects, to those that use a differential equation treatment. In this dissertation we investigate the spatio-temporal correlations properties associated to a stochastic model of agents that incorporates characteristics of heterogeneity and adaptability to a closed system in which agents are characterized by their income mi, expenditure rate !i and mobility i. A fraction q of the population has a spending rate positively correlated with their income while the remaining population correlates negatively. A second parameter p specifies the fraction of agents whose mobility may be correlated with the level of income, according to 4 possible scenarios. The variation of parameters (p; q) produces behaviors limiting the average expenditure rate and Gini index compatible with those observed in real data. The radial distribution of resources m(r) around the highest income agent, as well as its variance m2(r) particular cases results of models already introduced in the literature. Finally, the temporal correlations point to different system relaxation dynamics dependent on the q parameter.