Analisis Kemiskinan di Sulawesi Selatan dengan Regresi Nonparametrik Berbasis B-Spline

<p><em>Poverty is one of the problems faced by Indonesia, including in the province of South Sulawesi. This study aims to identify and understand the complex relationship between factors influencing poverty in South Sulawesi using nonparametric B-Spline regression. The data used are secondary data obtained from the publication of the Central Statistics Agency in the form of Data and Information on Poverty in Districts/Cities in Indonesia in 2022. The variables used are the percentage of poor people as the dependent variable, and the percentage of per capita expenditure on food, poverty depth index, and poverty severity index as independent variables. The best B-Spline model was obtained using order 2 for each independent variable, and one knot for each independent variable at a certain point. This model provides a Generalized Cross-Validation (GCV) value of 10.199728. The results of the analysis show that the measure of the goodness of the model obtained or R<sup>2</sup> which means that the percentage of per capita expenditure on food, poverty depth index, and poverty severity index greatly influences the percentage of poor people in South Sulawesi. The relationship between the independent variables and the dependent variables is non-linear and varies. The B-Spline model can produce an accurate and flexible picture of the relationship pattern and variability in poverty data in South Sulawesi. This study can provide in-depth insights and recommendations for the government in the form of poverty alleviation policies based on local data and non-linear analysis, by targeting specific interventions according to the unique conditions of each region in South Sulawesi to increase the effectiveness of poverty alleviation.</em></p>