AN EXPLORATORY DATA ANALYSIS (EDA) APPROACH FOR ANALYSIS FINANCIAL STATEMENTS IN PHARMACEUTICAL COMPANIES USING MACHINE LEARNING

• Main Author: Mega Panji Santosa, Cahya
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/83978
• Institution: Institut Teknologi Bandung

This research investigates the use of Exploratory Data Analysis (EDA) and machine learning techniques to analyze financial statements of pharmaceutical companies. The study focuses on three major Indonesian pharmaceutical companies: Kimia Farma, Kalbe Farma, and Indofarma. Through the use of Exploratory Data Analysis (EDA), the study seeks to identify hidden patterns and valuable insights within their financial data, focusing on metrics such as earnings per share (EPS), return on capital employed (ROCE), net profit margin, and inventory turnover ratio. Moreover, the study incorporates various machine learning models, including Linear Regression, K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Decision Tree, to forecast financial performance metrics and trends. The models' performance was assessed using several metrics, including Root Mean Squared Error (RMSE), Mean Squared Error (MSE), Mean Absolute Error (MAE), and Mean Absolute Percentage Error (MAPE). Among all the models evaluated, the Decision Tree model stood out with the best performance, achieving an Rsquared value of 0.998, MAPE of 4.8%, MAE of 4.8 x 1010, and MSE of 7.76 x 1021. These results underscore the high accuracy and excellent fit of the Decision Tree model to the data, demonstrating the significant potential of data-driven approaches to enhance the operational efficiency and financial stability of healthcare organizations.