Agora exibindo 1 - 2 de 2
ItemCore GDP: a medium- to long-run measure of Brazilian economic activity(2018) Pimentel, Luana Moreira de Miranda; Luquett, Ingrid; Ferreira, Pedro CostaThe economic activity of a country can be affected by shocks due to short-duration events that do not represent change in the business cycle. Knowledge about the current state of the economy is useful in decision making and it is helpful to generate investment, for example. We apply the methodology developed by Altissimo et al. (2010) to the Brazilian scenario, aiming to construct an indicator of the state of the economy, the Core GDP, that is less affected by short-run fluctuations and is capable of predicting turning points in domestic economic growth. The article contributes to the literature since our indicator has all the desired properties of a core: low volatility, and the ability to capture and predict a trend in economic activity. The Core GDP proved to be effective in detecting signs of relevant changes in the Brazilian economic activity tendency, such as the pick-up in growth after the 2008 crisis, the economic retraction that started in 2014 and the beginning of a recovery in 2016. ItemMachine-learning techniques and short-term combination forecasting of industrial production(2018) Ferreira, MarcolinoThe aim of this study was to develop short-term forecasts of the industrial production index in Brazil. Forecasts are made using five different methodologies: SARIMA, regressions, a structural, a dynamic factor models and decision trees. The random forest method had the best accuracy and was markedly superior to the other techniques. The univariate models had the worst performance during the period studied. Forecast combination was effective in reducing the one-step-ahead error. For the month-overmonth variation, for example, the RMSE, which varied between 1.27 and 7.57 for the individual models, was reduced to 0.85 for one of the combinations.