Stock Market Trend Prediction Using High-Order Information of Time Series

Abstract

Given a financial time series such as S&P 500, or any historical data in stock markets, how can we obtain useful information from recent transaction data to predict the ups and downs at the next moment? Recent work on this issue shows initial evidence that machine learning techniques are capable of identifying (non-linear) dependency in the stock market price sequences. However, due to the high volatility and non-stationary nature of the stock market, forecasting the trend of a financial time series remains a big challenge. In this paper, we introduced a new method to simplify noisy-filled financial temporal series via sequence reconstruction by leveraging motifs (frequent patterns), and then utilize a convolutional neural network to capture spatial structure of time series. The experimental results show the efficiency of our proposed method in feature learning and outperformance with 4%-7% accuracy improvement compared with the traditional signal process methods and frequency trading patterns modeling approach with deep learning in stock trend prediction. INDEX TERMS Trend prediction, convolutional neural network, financial time series, motif extraction.

Cross Entropy Based Long Short Term Memory Recurrent Neural Network Model for Analyzing the Time Series on Stock Market Price

“…M. Wen, P. Li, L. Zhang, and Y. Chen [7] developed a novel technique for predicting time series in the field of finance by reconstructing time series through higher order structures. The data set included in the proposed method are the Standard & Poor's 500 stock (S&P 500) index and the individual stocks like Google Inc. (GOOGL), International Business Machines Corporation (IBM), The Boeing Company (BA), etc.…”

“…The daily information of each sample includes high price, low price, close price, opening price, and trading volume. The stock price of apple generated each week from the first week of September 2007 to the end of August 2012 is extracted from the Yahoo finance website [12]. This data set contains the high, open, close, low and adjusted close prices of apple stock on every Monday throughout (2007)(2008)(2009)(2010)(2011)(2012).…”

“…4 shows the comparison of the graphs for the proposed method and the existing methods. The values obtained from T. Fischer and C. Krauss [11] for LSTM and Wavelet techniques evaluated by C. Stolojescu, A. Cusnir, S. Moga, and A. Isar [12] are compared with the proposed LSTM-RNN model, with various financial markets. The developed models used a number of features that caused gradient vanishing and data redundancy.…”

Predicting Stock Market Trends using Hybrid SVM Model and LSTM with Sentiment Determination using Natural Language Processing

“…In the field of Machine Learning (ML), the most popular algorithm used is the Support Vector Machine (SVM), [1] the SVM has been widely used to predict the stock market trends since it is noise-tolerant and gives a decent accuracy. Other ML algorithm used is the K-Nearest Neighbour (KNN), [2] KNN has been used to predict stock market data to some extent but not with great accuracy. In the field of Deep Learning (DL), the most popular algorithms that have been used are the Artificial Neural Networks (ANN), [3] ANNs have been conventionally used to predict stock market trends but haven"t proved to be very efficient due to their limitation of data-overfitting.…”

Deep Learning for Stock Market Prediction

“…The study aimed to solve two problems, using CNNs and optimizing them for stock market data. Wen et al [ 28 ] applied the CNN algorithm on noisy temporal series by frequent patterns as a new method. The results proved that the method was adequately effective and outperformed traditional signal process methods with a 4 to 7% accuracy improvement.…”