Exploration of salience theory to deep learning: evidence from Chinese new energy market high-frequency trading

Abstract

Salience theory has been proposed as a new stock trading strategy. To assess the validity of this proposal, a complex decision trading system was constructed based on salience theory, a variational mode decomposition (VMD) model, a bidirectional gated recurrent unit (BiGRU) model, and high-frequency trading. The system selected 30 Chinese new energy concept stocks, ranked the stocks using salience theory, and selected the top and bottom three stocks for two portfolios. Twelve stages were established, following which the VMD and BiGRU models were applied to the predictions. The final predicted annualized returns for the high ST (salience theory value) group A (GA) and low ST group B (GB) were 194.06% and 165.88%, respectively. This finding validates the powerful utility of salience theory and deep learning to analyze the Chinese new energy market. Moreover, it explains the theoretical practicality issues that the short selling restriction is the essential reason, or even perhaps the only reason, that leads to the strength of salience theory.