<![CDATA[Stock is a promising investment and very flexible, investors can sell their stock at any time either part or all of their stock. The high potential yield offered is what makes the stock so famous among the investors. However, the level of participation of the Indonesian people in investing in the stock market is still very low. Fluctuating stock prices are also one of the factors people are reluctant to become stock investors. Therefore, investment in stocks requires a good analysis so that investors can increase profits, one of them is by predicting stock prices from time to time so that future stock prices can be predicted by conducting technical analysis. In this study, forecasting is done using one of the artificial neural network methods, namely Feedforward Neural Network (FFNN) which is trained using the Particle Swarm Optimization (PSO) method. PSO algorithm is considered capable of replacing the Backpropagation algorithm in training networks. The error rate from forecasting results is calculated using Mean Absolute Percentage Error (MAPE). In the test the smallest MAPE value is 1,793% and fitness value is 0.98239 with 4 input layers, 2 hidden layers, and 1 output layer on the network architecture.]]>
Copyrights © 2020
