The estimation and comparison in Turkey's export and import values of wood and wood products with using artificial neural networks and box-jenkins methods

In this study, we identified the most suitable time series and artificial neural network model with different architectures with using Box-Jenkins and Artificial Neural Networks (ANN) methods. The aim of these models can help to estimate and compare Turkey's import and export values of wood and forest products sector. The most appropriate ANN and Box-Jenkins models were estimated until 2016. The sub-sectors of the wood and wood product industry of import and export data (wood and wood products, paper and paper products and furniture industry) were analyzed separately. The data used in this study were obtained from Turkey Statistical Institute and monthly data covering the period of January 2003 and December 2014. The success of the estimated methods were compared using Root Mean Square Error, Mean Absolute Error and Mean Absolute Percentage Error Statistics. After the application, Box-Jenkins and Artifical Neural Network models have given successful results but Artifical Neural Network models have been found to have better prediction performance. Key Words: Wood and Wood Products Sector, Export, Import, Box- Jenkins Method, Artificial Neural Network