Employing AI to Predict the Cambodia Securities Exchange Index

This research examines the efficacy of linear regression and long short-term memory machine-learning models in forecasting the Cambodia Securities Exchange Index. The performance of each model is evaluated based on their root-mean-square errors, with a lower RMSE indicating better performance. The analysis utilizes daily data from February 1, 2019, to November 17, 2023, encompassing 1170 days. Out of these, 80% of the daily stock market index, which is equivalent to 936 days, is utilized for training the models. Conversely, the remaining 234 days are allocated for the test data. The comparison of linear and deep learning prediction of the daily stock index, based on the Test data and the root-mean-square error, reveals that RNN or LTSM stands out as the optimal prediction. Specifically, the test data reveal an estimated RMSE of 62.70 for linear regression, while the LSTM model achieves a lower error of 53.67. By examining the applicability of machine-learning algorithms in predicting the CSX Index, this study provides valuable insights. The given context underscores the exceptional performance of LSTM, ultimately concluding that AI techniques enhance the accuracy of predictions.