Desenvolvimento econômico regional: um modelo previsão nas Empresas Sergipanas

• Autor(a) principal: Silva, Luiz Inacio Sampaio da
• Data de Publicação: 2022
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFS
• Texto Completo: https://ri.ufs.br/jspui/handle/riufs/18763
• topic: Empresas; Entrada; Série temporal; Sergipe; Análise de dados; Region Development; Companies; Prediction; Time Series; CIENCIAS SOCIAIS APLICADAS::ECONOMIA

The actions of the development policy for the northeast have been gaining ground since the last century, in the construction of scenarios and models of economic development, thus allowing the elaboration of new businesses as a strategy to overcome the lack of development and that would lead to an attraction investment for all areas of activity. The development of this study began with the bibliographic review, presenting an economic panorama from the perspective of the developmentalists Myrdal, Hirschman and Perroux, with the support of books, dissertations, magazines and scientific articles, presented the expressiveness of economic development and the state of Sergipe in its definitions, characteristics and in the management of forecast information. In this way, with the intention of highlighting and presenting alternative models to explain the regional economic development of Sergipe, identifying the behavior of incoming companies, in order to verify possible situations that compromise the development, through the prediction and evolution in the next three years. The model developed serves as a monitoring of the scenario for private companies to invest in the future, as well as in the social context of the Sergipe population. The database comprises the period from 2008 to 2019, which required the extraction, organization of information and search for authors who have already presented studies on forecasting for the following years. Despite the few works related to regional development analysis and the perspective of predictive data science, we sought through the time series to describe the standard behavior of behavior and this was done through the visualization of graphs and tables. The best result identified ARIMA model was (0,1,2) with the lowest MAPE of 3.7180. For model validation, the year 2018 had the lowest percentage error of 0.206%, while for 2017 and 2019 it was 6.361% and 5.121%. Validation was carried out on the results that presented a better forecast for 2020, 2021 and 2022 with the respective results 5,661, 5,801 and 6,078 entering companies. The results addressed in the forecast model present the predictive validity for the proposed problem and a new knowledge of studies that ensure the strengthening of the develop mental economic in Sergipe.