Stock Market Prediction using Artificial Neural Network Backpropagation with Multivariate Regression.
Tendra Kristian, Farida Titik Kristianti
Telkom University
Abstract
Finance is highly nonlinear and completely random, a trading strategy is a fixed plan that is designed to achieve a profitable return by going long or short in markets. Stock market prices are highly unpredictable and volatile. This means that there are no consistent patterns in the data to create model stock prices over time near-perfectly. As technical indicators play important roles in building a strategy, we will use technical indicators such as momentum, volume and volatility as a variables input in Artificial Neural Network Backpropagation with Multivariate Regression. The models are evaluated using three statistical performance evaluation measures, Mean Squared Error (MSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). We choose the daily data on Indonesia Stock Echange (IDX High Divident-20 Stock Index) in 10 years of trading days and try to predict the daily closing price. Experimental results show that Artificial Neural Network Backpropagation with Multivariate Regression using 4 layer, ReLu activation, Adam optimizer L2 Regularization and Dropout can get a promising performance in the closing price prediction on the real data compared with other models.
Keywords: Stock Prediction, Multivariate Regression, Artificial Neural Network Backpropagation
Topic: Financial Technology