Use of Artificial Neural Networks and Arima to Forecasting Consumption Sawnwood of Pinus sp. in Brazil

Abstract

The objective of this study was to analyze the application of an artificial neural networks model and an ARIMA model to predict the consumption of sawnwood of pine. For this, we use real and secondary data collected and obtained from a historical data source, corresponding to the period from 1997 to 2016, which were later tested to generate the forecast models. Based on economic and statistical criteria, six explanatory variables were used to fit the best model. The choice of the model was made based on Mean Squared Error, Mean Absolute Error, Theil U metric, Percentage Error of Forecast and Akaike value information criterion. The results indicated that the models generated through the ARIMA model presented better performance when compared to the artificial neural network. The best adjusted model estimated a reduction of 1.33% in consumption of sawnwood of pine in Brazil for the period between 2017 and 2020.

Impacts of Global Climate Change on Duration of Logging Season in Siberian Boreal Forests

“…The methodology of forecasting of timber harvesting season duration dynamics is based on the well-known autoregressive-integrated-moving average modeling (ARIMA) framework suggested by G. Box and G. Jenkins in 1970 [57]. It is widely used in various disciplines, including some recent studies in forest economics [58][59][60].…”