Machine-learning techniques and short-term combination forecasting of industrial production

dc.contributorDemais unidades::RPCA
dc.creatorFerreira, Marcolino
dc.date.accessioned2020-06-19T20:42:26Z
dc.date.accessioned2022-11-03T20:36:56Z
dc.date.available2020-06-19T20:42:26Z
dc.date.available2022-11-03T20:36:56Z
dc.date.created2020-06-19T20:42:26Z
dc.date.issued2018
dc.identifierhttps://hdl.handle.net/10438/29319
dc.identifier.urihttps://repositorioslatinoamericanos.uchile.cl/handle/2250/5041690
dc.description.abstractThe 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.
dc.languageen_US
dc.subjectForecasting combination
dc.subjectMachine learning
dc.subjectIndustrial production
dc.subjectTime series
dc.subjectRandom forest
dc.subjectCombinação de previsões
dc.subjectProdução industrial
dc.subjectSéries temporais
dc.titleMachine-learning techniques and short-term combination forecasting of industrial production
dc.typePaper


Este ítem pertenece a la siguiente institución