Abstract: Stock market price data is huge and it changes every second. As it is a complex system in which people either make money or lose all their savings, hence it is important to understand the stock market. In the era of big and dynamic data, machine learning for predicting stock market prices and trends has become even more popular than ever. In this paper, we tried to predict the trend of the stock market. A model with a supervised machine learning algorithm is used to predict prices. We collected data of every company from the beginning from Yahoo finance and proposed comprehensive customization of RNN Machine Learning based models which are known as LSTM for predicting price trends of stock markets.
The proposed solution is comprehensive as it includes pre-processing of the stock market dataset, utilization of multiple feature engineering techniques, combined with an RNN based system for stock market price trend prediction.
In the yearly forecasting model, historical prices have been trained and achieved an accuracy of 84.0%.
We conducted comprehensive evaluations on frequently used machine learning models and concluded that our proposed solution outperforms due to the comprehensive feature engineering that we built.
Through our detailed design and evaluated prediction term lengths, feature engineering and data pre-processing methods, this work will help investors to invest in the stock by comparing stocks of different enterprises periodically, hence resulting in less risk. Also, it will contribute to the financial and technical domains of the stock analysis research community.
Keywords: Stock Market, Machine Learning, LSTM, RNN, Forecast, Feature Engineering
|
DOI:
10.17148/IJARCCE.2022.11339
