O objetivo deste trabalho é realizar previsões de séries de retornos de ações de empresas dos setores financeiro, de alimentos, industrial e de serviços, utilizando redes neurais artificiais (RNA) do tipo feedforward treinadas com algoritmo de Levenberg-Marquardt e modelos Arima-Garch. Selecionaram-se duas séries de cada setor, e os dados foram obtidos da economática. Para o setor financeiro, são analisadas as séries dos bancos Bradesco e Itaú, no setor de alimentos a Perdigão e a Sadia, no setor industrial a Marcopolo e a Gerdau, e no setor de serviços o Pão de Açúcar e Lojas Americanas. Verificou-se que as previsões realizadas pelas duas técnicas têm desempenhos parecidos, não revelando superioridade de nenhuma técnica.
The subject of study in this article is the relationship between human development, represented by the HDI, and the amounts of tax collected in each Federal State of Brazil, considering the taxes in the three spheres of government, namely the Union, the States and the Municipalities. The research utilized data available on the websites of the National Treasury, Federal Revenue, IBGE and Atlas Brazil in 2013 to build a statistical model correlating the HDI value with the total tax collected per capita of each state. From this regression function, it was possible to calculate the HDI expected values and to confront with the actually obtained by each State. Sorting the differences between these two values, it was set up a ranking that primarily seeks to compare the return of well-being reached by the States in contrast to the tax burden imposed on its population.
-Context. The principles behind the process of creating new, spontaneous sequences out of previously ordered non-declarative stimuli have been scarcely addressed and, for such reason, remain highly unknown. Objective. This paper has four interconnected goals: (1) Introduce a new software-based neuropsychological test that can be used as a mean to assess key aspects of the way people order and reorder non-declarative stimuli, based upon cognitive dissonance principles; (2) introduce a mathematical approach to the latter in ordering/reordering of non-declarative stimuli; (3) assess whether the principles of cognitive dissonance in ordering/re-ordering hold for a cohort of young adults with upper socio-economic level; (4) access the extent to which the same holds for children and adolescents and trace a curve of maturation of cognitive dissonance in ordering/re-ordering. Methods. Our multi-age and multi-language social Network Test implies the two stages, first the subject must order figures of human faces in order of preference, next, the software provides him with different pairs of figures which the subject must fulfill in order to built the intermediate arrays that he believe to interconnect the original pair. Our mathematical model is centered around the relation defined by increases in the distance separating these different pairs of figures in the initial order (distances 1, 5 and 11) and related increases in the mean number of intermediate arrays placed in the re-ordering phase; 105 subjects were tested. Results. The tendency to produce reorders that are consonant to the one produced in the initial phase increases with age. This trend inspired us to propose a cognitive dissonance index in spontaneous ordering/reordering of non-declarative stimuli, which may formalize the operation of a previously unknown cognitive dimension of the human mind and may serve as an index of cognitive maturation. To the extent that further studies endorse these perspectives, the tests, formulas, and theoretical principals may support new diagnostic methods and explorations in cognitive science.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.