Dual-CLVSA: a Novel Deep Learning Approach to Predict Financial Markets with Sentiment Measurements
This work addresses the challenge of financial market prediction for investors by incorporating sentiment data, though it is incremental as it builds on an existing CLVSA model.
The authors tackled the problem of predicting financial market movements by integrating trading data with social sentiment measurements using a novel deep learning model, dual-CLVSA, which achieved effective data fusion and demonstrated that sentiment measurements provide extra profitable features to boost prediction performance, as validated through backtesting on SPDR SP 500 Trust ETF data over eight years.
It is a challenging task to predict financial markets. The complexity of this task is mainly due to the interaction between financial markets and market participants, who are not able to keep rational all the time, and often affected by emotions such as fear and ecstasy. Based on the state-of-the-art approach particularly for financial market predictions, a hybrid convolutional LSTM Based variational sequence-to-sequence model with attention (CLVSA), we propose a novel deep learning approach, named dual-CLVSA, to predict financial market movement with both trading data and the corresponding social sentiment measurements, each through a separate sequence-to-sequence channel. We evaluate the performance of our approach with backtesting on historical trading data of SPDR SP 500 Trust ETF over eight years. The experiment results show that dual-CLVSA can effectively fuse the two types of data, and verify that sentiment measurements are not only informative for financial market predictions, but they also contain extra profitable features to boost the performance of our predicting system.