การพยากรณ์และการวิเคราะห์ตัวแปรทางด้านเศรษฐกิจ ฐานความรู้ที่มีผลต่อการเจริญเติบโตของผลิตภัณฑ์มวลรวมประเทศโดยใช้เทคนิคด้านนิวโรฟัซซี

Abstract

The objectives of this independent study consisted of two areas. Firstly, to select appropriate variables of the knowledge-based economy indicators as alternative indicators for predicting Gross Domestic Product (GDP) growth. Secondly, to develop models for forecasting the GDP growth rate using neuro-fuzzy technique and compare the model performance. The data used in this work were collected from the World Bank through an Application Programming Interface, consisting of 5 regions: East Asia & Pacific, Europe & Central Asia, Latin America & Caribbean, Middle East & North Africa, and South Asia. The study investigated and identified the independent variables of the knowledge-based economy that could be used in the GDP growth rate prediction model along with the development of the Adaptive Neuro-fuzzy Inference System (ANFIS) to predict the GDP growth rate. The performance assessment used the prediction results to compare with the Linear Regression (LR) and Artificial Neural Network (ANN) models for accuracy, using the Mean Absolute Error (MAE) and Root Mean Square Error (RMSE). The result showed that the knowledge-based economy variables in this study have differing correlation significancy with the GDP growth. Particularly, the knowledge-based economy variables demonstrated the most significant relationship in the East Asia & Pacific and Latin America & Caribbean regions, respectively, while the variables showed weak relationship with the GDP growth in the Middle East & North Africa region. In the case of the development of ANFIS from the selected variables, the ANFIS provided the highest accuracy in predicting GDP growth in 14 of 15 experiments from three types of datasets: training dataset, testing dataset, and unseen dataset, while the ANN and LR models are less accurate, respectively. Also, the East Asia & Pacific region has the lowest error of all regions; with the average MAE and RMSE of the testing and unseen datasets at 0.265% and 0.345%, respectively. In addition, the developed ANFIS model can be good predictors in some regions, especially in the East Asia & Pacific, South Asia and Europe & Central Asia regions, while in the Middle East & North Africa, the error from the developed model is the highest one since the relationship between the selected input variables and output variable are very low.