Python-based stock price prediction using backpropagation neural networks: a case study on ANTM

Abstract

Accurate stock price prediction is critical for informed investment decisions. Today, stock trading has become a popular option as a source of income among people, due to its potential for rapid gains in a short time, but, due to fluctuating stock prices, it can cause great losses in exchange. This study aims to forecast the closing price using the backpropagation neural network algorithm so that it can be used as a decision support for potential investors and traders in this research, the system was built using the Python programming language, and the stock price data used were shares of the company Aneka Tambang Tbk (ANTM). The results of this research are root mean squared error (RMSE) values, additional labels for prediction results, and graphs for comparison of the original data with the predicted data. Based on the testing result, the best value of RMSE is 3.786, the mean absolute percentage error (MAPE) value is 0.001 which indicates that the prediction results are very close to the actual value.