<![CDATA[The Geometric brownian motion (GBM) model is widely used in predicting financial instruments such as stocks, because it can overcome the weakness of Brownian motion (BM) which can produce negative values. This study aims to apply the GBM model to predict the daily stock price of PT Bank Negara Indonesia (Persero) Tbk (BBNI) for the period from January to December 2024. The data used is secondary data on daily closing prices obtained from Investing.com, with a distribution of 95% training data and 5% testing data. Parameter drift and volatility are estimated using the Maximum Likelihood Estimation (MLE) method, while model accuracy is evaluated using MAPE and RMSE. The results show that a data proportion of 95%:5% provides the best prediction performance with a MAPE value of 5.724% and an RMSE of 0.267, indicating a high level of accuracy. Thus, the GBM model is reliable enough to describe the price movements of BBNI shares. Future research could develop models that take external factors into account or compare them with other stochastic models.]]>
Copyrights © 2026
