Are you interested in stock market? Here is some data on the future stock market by a leading news channel CNN.
Stock Market Predictions
Feature extraction from financial data is one of the most important problems in market prediction domain for which many approaches have been suggested. Among other modern tools, convolutional neural networks (CNN) have recently been applied for automatic feature selection and market prediction. However, in experiments reported so far, less attention has been paid to the correlation among different markets as a possible source of information for extracting features. In this paper, we suggest a CNN-based framework, that can be applied on a collection of data from a variety of sources, including different markets, in order to extract features for predicting the future of those markets. The suggested framework has been applied for predicting the next day’s direction of movement for the indices of S&P 500, NASDAQ, DJI, NYSE, and RUSSELL based on various sets of initial variables. The evaluations show a significant improvement in prediction’s performance compared to the state of the art baseline algorithms.
Financial markets are considered as the heart of the world’s economy inwhich billions of dollars are traded every day. Clearly, a good prediction offuture behavior of markets would be extremely valuable in various areas. Stockmarkets play an important role in Economic growth, (Beck & Levine, 2004)5so, analyzing their behavior and predicting their future can be very helpful inachieving economic goals. Another application of stock market prediction can befound in stock market trading systems, that usually consist of several modulesfor prediction, risk analysis and trading strategy. The goal of a trading moduleis to create a portfolio of stocks that maximizes the overall return regarding the10risk of stocks in that portfolio (Markowitz, 1952). However, a prediction modulefocuses on the sub-problem of predicting the future of the markets that can bea very valuable piece of information in the process of stock trading. So, theperformance of this module and by extent the whole trading system is inﬂuencedconsiderably by the quality of predictions that happen in the prediction module.15In fact, without a reliable prediction, it is almost impossible to have an excellenttrading system.Machine learning techniques have proved to be useful for making such predic-tions. Artiﬁcial neural networks (ANN) and support vector machine (SVM) arethe most common algorithms that have been utilized for this purpose (Guresen20et al., 2011; Kara et al., 2011; Wang & Wang, 2015). Statistical methods, ran-dom forests (Khaidem et al., 2016), linear discriminant analysis, quadratic dis-criminant analysis, logistic regression and evolutionary computing algorithms,especially genetic algorithm (GA), (Hu et al., 2015b; Brown et al., 2013; Huet al., 2015a; Atsalakis & Valavanis, 2009) are among other tools and techniques25that have been applied for feature extraction from raw ﬁnancial data and/ormaking predictions based on a set of variables (Ou & Wang, 2009; Ballingset al., 2015).Deep learning (DL) is a class of modern tools that is suitable for automaticfeatures extraction and prediction (LeCun et al., 2015). In many domains,302 such as machine vision and natural language processing, DL methods have been shown that are able to gradually construct useful complex features from raw dataor simpler variables (He et al., 2016; LeCun et al., 2015). Since the behavior of stock markets is complex, nonlinear and noisy, extracting features that are informative enough for making predictions is a core challenge, and DL seems to35be a promising approach to that. Algorithms like deep multilayer perceptron(MLP) (Yong et al., 2017), restricted Boltzmann machine (RBM) (Cai et al.,2012; Zhu et al., 2014), long short-term memory (LSTM) (Chen et al., 2015;Fischer & Krauss, 2018), autoencoder (AE) (Bao et al., 2017) and convolutionalneural network (CNN) (Gunduz et al., 2017; Di Persio & Honchar, 2016) are40famous deep learning algorithms utilized to predict stock markets. It is important to pay attention to the diversity of the variables that can be used for making predictions. The raw price data, technical indicators which come out of historical data, other markets with a connection to the target mar-ket, exchange rates of currencies, oil price and many other variables can be useful45for a market prediction task. Unfortunately, it usually is not a straight forward task to aggregate such a diverse set of information in a way that an automatic market prediction algorithm can use them. So, most of the existing works in this ﬁeld have limited themselves to a set of technical indicators representing a single market’s recent history.