Physics of Earth and Complex Systems, Institute of Technology Bandung
Abstract
Econophysics is a discipline that applies various models and concepts derived from physics to economic and financial phenomena. One of the things studied in econophysics is stock price prediction. Stock prices change continuously according to market conditions. Investors can benefit through the difference between the selling price and the buying price that occurs. This makes stock price analysis and prediction important for making profits and avoiding losses when investing in stocks. One method that can be used to predict stock prices is the Lorentzian Distance Classifier (LDC). The LDC method is a Machine Learning classification algorithm capable of categorizing historical data from a multidimensional feature space. This study aims to examine the performance of the Lorentzian Distance Classifier (LDC) in predicting stock closing prices of TLKM. The results of this study conclude that the use of LDC increases the accuracy value which is better than conventional methods based on MAE, MAPE and R2 values.
