<![CDATA[Indonesia has the world's largest nickel resources, with production of 1.6 million tons out of a global total of 328 million tons by 2022. In 2020, the Indonesian government imposed a ban on nickel ore exports to increase domestic processing and attract investment. Nickel supply reached 26 billion tons with reserves of 11,887 million tons. Mineral and coal investment in 2021 reached US$35 billion. The government plans 53 smelters until 2024, with 19 operating in 2021. PT Resource Alam Indonesia Tbk is active in the industry and faces fluctuations in nickel stock prices, which create problems, namely uncertainty for investors in making investment decisions due to fluctuations in nickel prices on the world market. So, effective stock price forecasting is needed using time series data analysis. This research uses a deep learning algorithm approach: Long Short Term Memory (LSTM). The research method uses CRISP-DM, which includes business understanding, data understanding, data preparation, model building, model evaluation, and deployment. Experimentation uses Python, and visualization uses the Streamlit Framework. This study uses optimal technical parameters to evaluate the LSTM model's effectiveness in predicting Nickel stock prices at PT Resource Alam Indonesia Tbk. The results showed that the Long Short Term Memory (LSTM) model could predict the sale of Nickel shares at PT. Resource Alam Indonesia Tbk (password: KKGI.JK) well, with an MAE value of 33.15, RMSE value of 48.14, MSE value of 2317.33, and MAPE value of 7.39. The best combination of the parameter combinations tested is with batch size 32, epochs 150, and optimizer Adam. The findings provide valuable insights for investors in making more informative and effective investment decisions.]]>
