Abstract
The stock market is known for its extreme complexity and volatility, and people are always looking for an accurate and effective way to guide stock trading. Long short-term memory (LSTM) neural networks are developed by recurrent neural networks (RNN) and have significant application value in many fields. In addition, LSTM avoids long-term dependence issues due to its unique storage unit structure, and it helps predict financial time series. Based on LSTM and an attention mechanism, a wavelet transform is used to denoise historical stock data, extract and train its features, and establish the prediction model of a stock price. We compared the results with the other three models, including the LSTM model, the LSTM model with wavelet denoising and the gated recurrent unit(GRU) neural network model on S&P 500, DJIA, HSI datasets. Results from experiments on the S&P 500 and DJIA datasets show that the coefficient of determination of the attention-based LSTM model is both higher than 0.94, and the mean square error of our model is both lower than 0.05.
Highlights
Financial market forecasting has traditionally been a focus of industry and academia.[1]
Long short-term memory (LSTM) neural networks have performed well in speech recognition[3, 4] and text processing.[5, 6]. Because they have the characteristics of selectivity, memory cells, LSTM neural networks are suitable for random nonstationary sequences such as stock-price time series
The results show that the deep and wide area neural network (DWNN) model can reduce the predicted mean square error by 30% compared to the general recurrent neural networks (RNN) model
Summary
Financial market forecasting has traditionally been a focus of industry and academia.[1]. A hybrid model of generalized autoregressive conditional heteroskedasticity (GARCH) combined with LSTM was proposed to predict stock price fluctuations. LSTM uses one of the most common forms of RNN.[17] This time recurrent neural network is meant to avoid long-term dependence problems and is suitable for processing and predicting time series.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
