Employing Deep Learning In Intraday Stock Trading

Abstract

Accurate stock price prediction is a significant benefit to the Stock investors. The future Stock value of any company is determined by Stock market prediction. A successful prediction of the stock’s future price could result in a significant profit; Hence investors prefer a precise Stock price prediction. Although there are many different approaches to helps in forecasting stock prices, this paper will briefly look into the deep learning models and compare LSTM model and its variants. The key intention of this study is to propose a model that is best suitable and can be implemented to forecasting trend of stock prices. This paper focuses on binary classification problem, predicting the next-minute price movement of SPDR S&P 500 index The testing experiments performed on the SPDR S&P 500 index reveals that the variants of LSTM models, Slim LSTM1, slim LSTM2, and Slim LSTM3 with less parameters, provide better performance when compared to the Standard LSTM Model.