Browsing by Author "Nurleta, Evi"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    PENERAPAN METODE EXTREME GRADIENT BOOSTING PADA KLASIFIKASI STATUS KEBANGKRUTAN PERSEROAN TERBATAS DI BURSA EFEK INDONESIA
    (Elfitra, 2023-07) Nurleta, Evi; Harison, Harison
    Corporate bankruptcy is characterized by financial difficulties. To classify with the aim of predicting the bankruptcy status of Limited Liability Companies on the Indonesia Stock Exchange in 2021, it can be done using the XGBoost method. The variables used are the financial ratios Net Working Capital to Total Assets (WCTA), Retained Earnings to Total Assets (RETA), Earnings Before Interest and Tax to Total Assets (EBITTA), Book Value of Equity to Book of Debt (BVETA). The bankruptcy status is safe, gray (needs special attention), and distress (bankrupt). Based on the classification results using training data of 80% and testing data of 20%, a classification accuracy rate of 93.00%, 90.95% precision and 90.28% sensitivity is obtained, and it is known that the WCTA and RETA variables are the most important features in classifying bankruptcy classes. The results of this classification can be used for forecasting the bankruptcy of a Limited Liability Company and as a reference for making policies in dealing with bankruptcy.