Predicting the price of crude palm (CPO) oil is vital for resources management, especially in agricultural farms. However, the price of CPO is very volatile in uncertain economic conditions and the agricultural environment. In addition to this volatility, the CPO price presents non-linearity features, making its prediction challenging. The authors present a deep learning approach for the CPO price prediction. The researchers compare a SARIMA model with three deep learning techniques: Multilayer Perceptron, Long Short Term Memory (LSTM), and Simple Recurrent Neural Network to uncover the most accurate prediction model for the CPO prices. The results suggest that the LSTM-based modeling approach presented in this research outperformed their counterparts in predicting the CPO price in terms of prediction accuracy. The findings suggest that the proposed LSTM based forecasting approach is a useful and reliable deep learning technique that may provide valuable information to businesses, industries, and government agencies.