Admissão e demissão: uma análise multivariada de curto e longo prazo sob a ótica do setor eletroeletrônico brasileiro

Abstract

The objective of this research was to verify if the Cluster analysis would be able to classify and select the macroeconomic variables related to the Brazilian electroelectronic sector in exogeneity terms, as well as to adjust a Vector Autoregressive Model (VAR) and a Vector Error Correction (VEC) which point out the short- and longterm relationship between the variables. The time series that were part of this study were: admission and dismissal in the Brazilian electronics industry, Gross Domestic Product, total Brazilian import and export, monthly average dollar quotation, national broad consumer index and settlement and custody system. The sample period comprised May of 2003 to February of 2017 for the first article, and May of 2003 to September of 2018 in the case of the second article. The variables of the sector were collected in the online platform of the Brazilian Association of the Electrical and Electroelectronic Industry (Abinee), and the others, Ministry of Labor and Employment, General Register, Central Bank of Brazil, Rotary Brazil; Ministry of Finance, and the Brazilian Institute of Geography and Statistics. At the end of the study Cluster analysis proved to be efficient in the classification and selection of variables considering exogeneity, being corroborated by the Granger / Block Exogeneity Causality Test. It was evident that there is a significant relationship between them, indicating that abrupt changes in the macroeconomic variables will result in fluctuations in the number of employees of the Brazilian electrical and electronic industry by at least two months.