Diversification of agricultural areas in Indonesia using dynamic copula modeling and K-Means clustering

Abstract

Agriculture is one of the main pillars of economic growth in Indonesia. Failure in this sector can result in faltering economic stability of the country. Thus, to minimize these failures, mapping of areas with particular commodity potential is needed. One of the main factors affecting the growth of crops is rainfall. Therefore, this paper aims to model the potential distribution of commodity growth based on rainfall precipitation using dynamic copula. The modeling results are then used as a basis for grouping the potential of food crop commodities in Indonesia. The determination of the group was carried out using the k-means clustering method. We expect that the result of the modeling can provide an overview for farmers or the government to make policies related to the optimization of Indonesia's agricultural sector. This result will enable the government to offer facilities that can minimize agricultural losses, such as superior seeds that are resistant to weather changes and the provision of training for enhancing farming skills. In addition, it is also suggested to diversify farm areas to reduce the failures due to dependence on a single agricultural product.