This is the authors' abstract. We don't add key points for this paper.
This study compared autoregressive integrated moving average (ARIMA) models and artificial neural networks (ANNs) in analysing Brazil’s forest product production time series. Data on charcoal, pulp and composite wood panel production from 1961 to 2019 was sourced from the FAO database. Model performance was assessed using root mean squared error, mean absolute error and U-Theil statistics. The ARIMA model and ANNs showed similar predictive accuracy, however ARIMA delivered more consistent projections considering the historical trends of the analysed time series. Charcoal production has shown high variability in recent years, with expectations of stabilisation or decline due to COVID-19 impacts. Projections for pulp and wood panel production suggest potential growth, supported by Brazil’s competitive advantages. However, the possibility of reduced forest product output due to global economic downturns must be considered.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Pereira et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: