@phdthesis{digilib63758,
           month = {January},
           title = {ROBUST PRINCIPAL COMPONENT REGRESSION DENGAN METODE MINIMUM COVARIANCE DETERMINANT (MCD) DAN MINIMUM VOLUME ELLIPSOID (MVE) MENGGUNAKAN ESTIMATOR LEAST TRIMMED SQUARE (LTS)  (STUDI KASUS: DATA KEMISKINAN INDONESIA MENURUT PROVINSI PADA TAHUN 2022)},
          school = {UIN SUNAN KALIJAGA YOGYAKARTA},
          author = {NIM.: 20106010021 Arditya Criszardin},
            year = {2024},
            note = {Pembimbing: Mohammad Farhan Qudratullah, S.Si., M.Si},
        keywords = {Kemiskinan, Robust Principal Component Regression, Minimum Covariance Determinant, Minimum Volume Ellipsoid, Outlier},
             url = {https://digilib.uin-suka.ac.id/id/eprint/63758/},
        abstract = {Poverty in Indonesia is evenly distributed across regions; however, its severity varies in each region. According to data from the Central Statistics Agency (BPS) in September 2022, the number of poor people reached 26.36 million, an increase of 0.20 million from March 2022 and a decrease of 0.14 million compared to September 2021. The growth of poverty occurs both in urban and rural areas. Therefore, the robust principal component regression approach is employed to analyze the factors influencing poverty in Indonesia. Based on previous literature, influencing factors include the open unemployment rate, regional gross domestic product, poverty severity index, average years of schooling, human development index, and regional minimum wage. The results indicate that the robust principal component regression model using the minimum covariance determinant method with the least trimmed square estimator performs better, with a residual standard error of 71.54606 compared to the minimum volume ellipsoid method.}
}