论文信息 - Comparison of Predictive Algorithms: Backpropagation, SVM, LSTM and Kalman Filter for Stock Market

Comparison of Predictive Algorithms: Backpropagation, SVM, LSTM and Kalman Filter for Stock Market

The variation and dependency on different parameters of stock market makes prediction a complex process. Artificial neural Networks have been proven to be useful in such cases to predict the stock values. The parameters involved and the commonly used algorithms are discussed and compared in this paper. In case of backpropagation algorithm, a feed forward network is present and weights are modified by back propagating the error. Similarly, significant modification is introduced in Sup-port Vector Machines Algorithm(SVMA) which results in higher accuracy rates. Presence of kernel and other parameters make it more flexible. Long Short-Term Memory(LSTM), another commonly used time series forecasting algorithm, is a special type of Recurrent Neural Network(RNN) that uses gradient descent algorithm. This paper provides a comparative analysis between these algorithms on the basis of accuracy, variation and time required for different number of epochs. The T-test hypothesis test was used for further analysis to test the reliability of each algorithm.

[1] Gaël Varoquaux,et al. Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[2] Ganesh Bonde. Extracting the best features for predicting stock prices using machine learning , 2012 .

[3] Swati Bhatt,et al. Predicting Stock Price Direction using Support Vector Machines , 2015 .

[4] Tony Lacey,et al. Tutorial: the Kalman Filter 11.1 Introduction 11.2 Mean Squared Error , 1998 .

[5] Alice Zheng. Using AI to Make Predictions on Stock Market , 2017 .

[6] Omar S. Soliman,et al. A Machine Learning Model for Stock Market Prediction , 2014, ArXiv.

[7] Xu Yan,et al. Application of Kalman Filter in the Prediction of Stock Price , 2015 .

[8] Bernhard Schölkopf,et al. A tutorial on support vector regression , 2004, Stat. Comput..

[9] Yulong Liu,et al. Support Vector Regression for prediction of stock trend , 2013, 2013 6th International Conference on Information Management, Innovation Management and Industrial Engineering.