Share Price Prediction for Increasing Market Efficiency using Random Forest

Abstract

The price of a single share of a collection of sell-able shares, options, or other financial assets, shall be the price of a share price. The share price is unpredictable since it primarily depends on buyers’ and sellers’ expectations. Share is a primary and secondary market equity security. In this study we will use machine learning techniques to predict the share price for increasing market efficiency. In addition, it is important for us to build a models to create appropriate features to improve the performance of the models. The random forest and the recurrent neural network will be used to achieve this. To fix class imbalance, we analyse preprocessing of the data set, like the selection of the features using filter and wrapper methods and selected oversampling techniques. The model’s performance will be evaluated using Mean absolute error (MAE), Mean square error (MSE), Root mean square error (RMSE), Relative MAE (rMAE), and Relative RMSE (rRMSE). The performance of the RNN and Rf algorithms was compared for the prediction of the closing price. The Rf model was found to be the best model for predicting the stock price (closing price). This research project together with its findings will have an impact in increasing market efficiency. This will also promote potential economic growth.