Comparison of Machine Learning Algorithms for Prediction of Total Assets

Assets are considered as a firm’s growth to determine firm’s future performance. The method for predicting future performance with total assets remains a gap for studying. Therefore, a comparative study is proposed in this paper to find out a predictive model for total asset prediction using different machine learning algorithms, which include Decision Trees (DCT), Gradient-Boosted Trees (GBT), K-Nearest Neighbors (KNN), and Random Forest (RF). The dataset used for the experiment comprises 1,224 samples, which is the statement of financial position of 204 companies from the year 2016 to 2021, gathered from the Stock Exchange of Thailand. The comparison results showed that GBT with splitting data into 80% for training and 20% for testing models performed the highest accuracy of 96.67%, compared to other algorithms.