Forecasting of Türkiye's industrial wood production via some trend analysis tools

This study deals with forecasting of Türkiye's industrial wood production and makes a projection for future five years. The data used within the scope of the study were provided from the statistics published by the General Directorate of Forestry, and in this regard, annual data covering the years 2000-2023 were used. The forecasting of industrial wood production was performed via the linear, exponential growth, quadratic and s-curve trend analysis functions. The mean absolute percentage error (MAPE) and the coefficient of determination (R2) were considered as the main performance evaluation criteria to compare the forecast success of the methods. Among the analysis methods considered, it was seen that the linear method yielded the highest MAPE with a value of 9.9238%, while the quadratic model yielded the lowest MAPE with a value of 6.2096%. Similarly, the linear model gave the lowest R2 with a value of 0.9098, while the quadratic model gave the highest R2 with a value of 0.9402. In summary, although the outcomes of all methods were satisfactory, it has been understood that the best outcomes in terms of performance criteria in forecasting of industrial wood production for the current data set were provided via the quadratic method.