A novel hybrid model based on recurrent neural networks for stock market timing
- PDF / 1,944,970 Bytes
- 18 Pages / 595.276 x 790.866 pts Page_size
- 19 Downloads / 191 Views
METHODOLOGIES AND APPLICATION
A novel hybrid model based on recurrent neural networks for stock market timing Yue Qiu1 · Hao-Yu Yang1 · Shan Lu1 · Wei Chen1
© Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Stock market timing is regarded as a challenging task of financial prediction. An accurate prediction of stock trend can yield great profits for investors. At present, recurrent neural networks (RNNs) have a good performance in stock market forecasting. However, there has been a relative lack of research in the stock market timing using RNNs. In this paper, a novel model named hybrid RNN model is proposed for stock market timing by incorporating multi-layer long short-term memory, multi-layer gated recurrent unit and one-layer ReLU layer. Moreover, based on five popular benchmark datasets from UCI Machine Learning Repository and six daily securities from Shanghai Stock Exchange, comparisons with 12 state-of-the-art models are conducted to verify the superiority of the proposed hybrid RNN model in terms of nine technical indicators. The findings from the experiment demonstrate that: (1) as opposed to 12 models, the average accuracy, MSE and AUC of hybrid RNN model (0.7406, 0.2592, 0.7368) are significantly better than other comparison models, and (2) the proposed hybrid RNN classification procedure can be considered as a feasible and effective tool for stock market timing. Keywords Stock market timing · RNN · LSTM · GRU · Classification · Deep learning
1 Introduction Stock market timing is valuable for financial experts and investors, because investing stock market rationally and choosing the right time can yield great profit and diversify portfolio risk. The efficient market hypothesis (EMH) is one of the major financial theories. EMH states that the mathematical expectation of stock market volatility is always zero. It means stock market timing cannot generate excess returns. Recently, several researches (Bollerslev et al. 2014; Kim et al. 2011; Phan et al. 2015) state that some markets such as emerging markets are worthy of being researched and explored. However, stock market timing is so difficult to make a accurate prediction. For example, there are a lot of factors to affect stock market’s performance including political events, legal factors, domestic situation, international situation, government policies and psychological factors (Majhi et al. 2014). In addition, government policies and legal forms Communicated by V. Loia.
B 1
Wei Chen [email protected] School of Management and Engineering, Capital University of Economics and Business, Beijing, China
have vital impact influences on the functioning of the stock market. Therefore, many researchers have attempted to propose different methods for stock market timing. In general, we can partition the methodologies into three groups technical analysis, fundamental analysis and machine learning techniques (Zhang et al. 2019). Fundamental analysis aims at analyzing the economical, industrial and financial situation, company managemen
Data Loading...
