Nonparametric Kernel Regression Analysis on the Relationship between Marine Capture Fisheries Production and the Price Index Received by Fishermen in the Arafura Sea

Abstract

The Arafura Sea is known as an area rich in fisheries resources, which is the main source of income for many local Fishermen. The main problem arises due to price and production fluctuations which can significantly affect Fishermen's income. Kernel nonparametric regression analysis was chosen as a research method to overcome the limitations of parametric regression models in capturing complex and nonlinear patterns in these relationships. This research aims to fill this knowledge gap by using a nonparametric kernel regression analysis approach. Thus, it is hoped that this research can provide new insights into the relationship patterns between marine capture fisheries production and price indices, opening the door to increasing the sustainability of the fisheries sector and the welfare of Fishermen in this region. This research is quantitative and uses nonparametric kernel regression data analysis to explore the relationship between marine capture fisheries production and the price index received by Fishermen in the Arafura Sea. Based on nonparametric kernel regression analysis with bandwidth selection methods, such as Cross-Validation, Generalized Cross-Validation, and Mean Squared Error, this research shows variations in estimates of marine capture fisheries production and the price index received by Fishermen in the Arafura Sea. Differences in estimation results between methods reflect different approaches in dealing with the trade-off between model fit to the data and model complexity. Although the results suggest flexible relationships, it is important to remember that conclusions about causal relationships cannot be drawn directly, and this study provides valuable insight into the complex dynamics of factors influencing fisheries production in the region.