A novel text mining approach to financial time series forecasting

By Baohua Wang, Hejiao Huang, Xiaolong Wang. 2012

127 citations recorded by the source corpus.

Financial time series forecasting has become a challenge because it is noisy, non-stationary and chaotic. Most of the existing forecasting models for this problem do not take market sentiment into consideration. To overcome this limitation, motivated by the fact that market sentiment contains some useful forecasting information, this paper uses textual information to aid the financial time series forecasting and presents a novel text mining approach via combining ARIMA and SVR (Support Vector Regression) to forecasting. The approach contains three steps: representing textual data as feature vectors, using ARIMA to analyze the linear part and developing a SVR model based only on textual feature vector to model the nonlinear part. To verify the effectiveness of the proposed approach, quarterly ROEs (Return of Equity) of six security companies are chosen as the forecasting targets. Comparing with some existing state-of-the-art models, the proposed approach gives superior results. It indicates that the proposed model that uses additional market sentiment provides a promising alternative to financial time series prediction.

Subfields: Artificial Intelligence, Management Science and Operations Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Qing Li
2. Yuanzhu Peter Chen
3. Yuntao Yao
4. Philip S. Yu
5. Yang Yang
6. Qixu Gong
7. Yuhong Liu
8. B. Fang

In Management Science and Operations Research

1. Qing Li
2. Yuanzhu Peter Chen
3. Ping Li
4. Qixu Gong
5. Tiejun Wang
6. LiLing Jiang
7. Ling Liu
8. Zhangxi Lin

Citations in the corpus

• Improving stock market prediction via heterogeneous information fusion (2017, methodology): “In [26], textual data are represented as feature vectors by using the bag-of-words approach, and the data are organized as the input of a time series model, showing the use of market sentiments can improve the forecasting accuracy.”
• The effect of news and public mood on stock movements (2014, methodology): “Notably, we assume zero transaction cost, as previously described [7,21,36,45].”
• A Tensor-Based Information Framework for Predicting the Stock Market (2016, methodology): “Wang et al. [2012] studied the relation between news and stocks using a hybrid predictive model based on both ARIMA and SVR. Bollen et al. [2011] measured public sentiment expressed in Twitter posts and used this indicator to forecast stock price trends with Self-Organizing Fusion Neural Networks…”
• Tensor-Based Learning for Predicting Stock Movements (2015, unknown): “…in previous studies (Lavrenko et al. 2000; Li et al. 2014a; Mittermayer and Knolmayer 2006; Schumaker and Chen 2008; 2009b; Schumaker et al. 2012; Wang, Huang, and Wang 2011; Xu and Zhang 2013; Yu, Duan, and Cao 2013) is to concatenate features of different information sources into one super…”
• Stock closing price prediction based on sentiment analysis and LSTM (2019, background): “However, stock market prediction is generally regarded as one of the most challenging problems in time-series prediction due to its characteristics of noise and volatility [2].”
• Media-aware quantitative trading based on public Web information (2014, methodology): “[40] proposed a hybrid stock prediction approach by combining ARIMA and SVR together.”