GDP Estimation and Slow Down Signal Model for Indonesia: An Artificial Neural Network Approach

Abstract

The purpose of this paper is to develop a model estimation of Gross Domestic Product (GDP) and economic slowdown signal models with an artificial neural net-work approach. This approach is as an al-ternative or complement to other ap-proaches that are widely used such as re-gression model. An artificial neural net-work model is inspired from the biological sciences such as the working of the hu-man brain in solving problems. In this study, external sectors have substantial role in influencing the growth of GDP. Almost 90 percent of leading indicators of external factors contributing the fluctuation of In-donesian GDP. Major trading partner of In-donesian manufactured goods such as China, South Korea, US and Japan to some extents, however, affect GDP fluctuation. The diversification of trading partners and commodities to be exported is one of the most important policies to reduce external shocks. Based on the model developed, the performance model is adequate in pre-dicting the samples – in and outside – in terms of a lower error. This model, how-ever, is still experimental in nature. There-fore, it needs to be further developed by using different topology and adding obser-vations.