Factors Affecting The Open Unemployment Rate in West Sumatra Province Using Spatial Autoregressive (SAR)

Abstract

This paper proposes a Spatial Autoregressive (SAR) model to analyze the significant factors affecting the open unemployment rate in West Sumatra during 2023. The main advantage of the method is its ability to accurately capture spatial interactions between neighboring regions, such that it can provide a comprehensive understanding of regional unemployment patterns efficiently. By introducing the K Nearest Neighbor (KNN) weighting matrix and spatial lag parameter to the model, the effect of regional proximity on unemployment rates is more accurately captured. The viability of the SAR model is assessed by analyzing its ability to produce the lowest Akaike’s Information Criterion (AIC) value, indicating its suitability for modeling regional unemployment patterns. The result indicates that the SAR model is more effective than the multiple linear regression model in capturing regional unemployment patterns, with an AIC value of 52.756. The factors that influence the open unemployment rate are gross regional domestic product, labor force participation rate and the percentage of poor people.