Stock Trend Prediction Research Articles

Forecasting stock market short-term trends using a neuro-fuzzy based methodology

A neuro-fuzzy system composed of an Adaptive Neuro Fuzzy Inference System (ANFIS) controller used to control the stock market process model, also identified using an adaptive neuro-fuzzy technique, is derived and evaluated for a variety of stocks. Obtained results challenge the weak form of the Efficient Market Hypothesis (EMH) by demonstrating much improved and better predictions, compared to other approaches, of short-term stock market trends, and in particular the next day’s trend of chosen stocks. The ANFIS controller and the stock market process model inputs are chosen based on a comparative study of fifteen different combinations of past stock prices performed to determine the stock market process model inputs that return the best stock trend prediction for the next day in terms of the minimum Root Mean Square Error (RMSE). Gaussian-2 shaped membership functions are chosen over bell shaped Gaussian and triangular ones to fuzzify the system inputs due to the lowest RMSE. Real case studies using data from emerging and well developed stock markets – the Athens and the New York Stock Exchange (NYSE) – to train and evaluate the proposed system illustrate that compared to the “buy and hold” strategy and several other reported methods, the proposed approach and the forecasting trade accuracy are by far superior.

Using support vector machine with a hybrid feature selection method to the stock trend prediction

In this paper, we developed a prediction model based on support vector machine (SVM) with a hybrid feature selection method to predict the trend of stock markets. This proposed hybrid feature selection method, named F- score and Supported Sequential Forward Search (F_SSFS), combines the advantages of filter methods and wrapper methods to select the optimal feature subset from original feature set. To evaluate the prediction accuracy of this SVM-based model combined with F_SSFS, we compare its performance with back-propagation neural network (BPNN) along with three commonly used feature selection methods including Information gain, Symmetrical uncertainty, and Correlation-based feature selection via paired t-test. The grid-search technique using 5-fold cross-validation is used to find out the best parameter value of kernel function of SVM. In this study, we show that SVM outperforms BPN to the problem of stock trend prediction. In addition, our experimental results show that the proposed SVM-based model combined with F_SSFS has the highest level of accuracies and generalization performance in comparison with the other three feature selection methods. With these results, we claim that SVM combined with F_SSFS can serve as a promising addition to the existing stock trend prediction methods.

Application of wrapper approach and composite classifier to the stock trend prediction

The research on the stock market prediction has been more popular in recent years. Numerous researchers tried to predict the immediate future stock prices or indices based on technical indices with various mathematical models and machine learning techniques such as artificial neural networks (ANN), support vector machines (SVM) and ARIMA models. Although some researches in the literature exhibit satisfactory prediction achievement when the average percentage error and root mean square error are used as the performance metrics, the prediction accuracy of whether stock market goes or down is seldom analyzed. This paper employs wrapper approach to select the optimal feature subset from original feature set composed of 23 technical indices and then uses voting scheme that combines different classification algorithms to predict the trend in Korea and Taiwan stock markets. Experimental result shows that wrapper approach can achieve better performance than the commonly used feature filters, such as χ 2-Statistic, Information gain, ReliefF, Symmetrical uncertainty and CFS. Moreover, the proposed voting scheme outperforms single classifier such as SVM, kth nearest neighbor, back-propagation neural network, decision tree, and logistic regression.

Multi-resolution subspace for financial trading

In this paper, we introduce a new stock trend prediction approach based on subspace classifier and a new feature representation. Our goal is not price prediction but rather trend prediction, which can be formulated as a problem of pattern classification. Recently, several works have approached this problem by applying machine learning techniques. We show that this problem con be efficiently solved using a new method of anchoring. From the feature extracted by technical indicators, we apply an anchoring method to create different features spaces, and a subspace classifier is trained in each feature space. A cascade of classifiers is developed to classify the patterns as “downward trend” or “upward trend”. Extensive experiments, carried out on various dataset, confirm the robustness of our approach. We show, that our method permit to obtain a gain higher than standard machine learning classifiers in all the tests.

Multiscale rough set data analysis with application to stock performance modeling

Multiscale Rough Set Data Analysis (MSRSDA) combines the feature selection and feature extraction ability of RSDA with the multiresolution property of the wavelet transform. RSDA removes the irrelevant (having no impact on prediction performance) and dependent (having no discriminatory power) market variables from the multivariate time series data matrix, thereby improving generalization capability of the forecasting engine (neural network, fuzzy inference system). A wavelet transform is used to decompose the time series into varying scales of resolution so that the underlying temporal structures of the original time series become more tractable; the decomposition is additive in details and approximation. Significance of market variables at various wavelet scales representing detailed short-term history and general long-term history of the time series, is computed using RSDA in wavelet space. Irrelevant and redundant variables at each scale are removed and a forecasting engine is then trained on each of the relevant resolution scales (i.e., those scales where significant events are detected); individual wavelets forecasts are recombined to form the overall forecast. The superior performance of MSRSDA is illustrated through stock performance modeling example. For prediction of S&P500 index, a data matrix of 21 variables has been used. MSRSDA gives 69% accuracy on stock-trend prediction.

Neural Networks as a Decision Maker for Stock Trading: A Technical Analysis Approach

There has been a growing interest in applying neural networks and technical analysis indicators for predicting future stock behavior. However, previous studies have not practically evaluated the predictive power of technical indicators by employing neural networks as a decision maker to uncover the underlying nonlinear pattern of these indicators. The objective of this paper is to investigate if using these indicators as the input variables to a neural network will provide more accurate stock trend predictions, and whether they will yield higher trading profits than the traditional technical indicators. Three neural networks are examined in the study to predict the short-term trend signals of three stocks across different market industries. The overall results indicate that the proportion of correct predictions and the profitability of stock trading guided by these neural networks are higher than those guided by their benchmarks.

Comparative study of stock trend prediction using time delay, recurrent and probabilistic neural networks

Three networks are compared for low false alarm stock trend predictions. Short-term trends, particularly attractive for neural network analysis, can be used profitably in scenarios such as option trading, but only with significant risk. Therefore, we focus on limiting false alarms, which improves the risk/reward ratio by preventing losses. To predict stock trends, we exploit time delay, recurrent, and probabilistic neural networks (TDNN, RNN, and PNN, respectively), utilizing conjugate gradient and multistream extended Kalman filter training for TDNN and RNN. We also discuss different predictability analysis techniques and perform an analysis of predictability based on a history of daily closing price. Our results indicate that all the networks are feasible, the primary preference being one of convenience.