Effective Stock Market Pricing Prediction Using Long Short Term Memory-Upgraded Model (LSTM-UP) on Evolving Data Sets
Tóm tắt
Making reliable stock market forecasts is a difficult real-world economics problem. A stock's unpredictable and chaotic nature makes it difficult to predict its future worth. To overcome the limitations of existing models in handling the non-stationary and non-linear characteristics of high-frequency financial time series data, this study proposes a Wavelet transform-based data preprocessing and the development of an LSTM-upgraded model (LSTM-UP) that incorporates human sentiment for predicting stock price. Features are extracted and trained using a Wavelet transform, long short-term memory, and an upgraded mechanism applied to financial time series. To investigate the role of human emotion, we added a sentiment polarity score to the raw data. Using the ADBL, NIB, NABIL, and SCB stock datasets, the suggested model is evaluated and compared to LSTM and GRU. Models' efficacy can be compared using metrics like root-mean-square error, mean absolute error, R2, and MDA. The Root-Mean-Squared Error (RMSE) and the Mean Absolute Error (MAE) were both less than 3.5 whereas R2 and MDA were both greater than 0.95 across all stock datasets used in the experiments. Adding human judgment to the model, as suggested, makes it superior than similar ones.
Tài liệu tham khảo
Refenes AN, Zapranis A, Francis G. Stock performance modeling using neural networks: a comparative study with regression models. Neural Net. 1994;7(2):375–88.
Vishwanath R, Krishnamurti C. Investment management: a modern guide to security analysis and stock selection. Berlin Heidelberg: Springer- Verlag; 2009.
H. G. J. H. Yuling Lin, "An SVM-based approach for stock market trend prediction," in Proceedings of the International Joint Conference on Neural Networks, Dallas, TX, 2013.
Bao W, Yue J, Rao Y. A deep learning framework for financial time series using stacked autoencoders and long short term memory. PLOS ONE. 2017;12:7.
Qin J, Wang B, Zhou C. Forecasting stock prices with long short term memory neural network based on attention mechanism. PLOS ONE. 2020;20:20.
Jin Z, Yang Y, Liu Y. Stock closing price prediction based on sentiment analysis and LSTM. Neural Comput Appli. 2019;32:9713.
Yahya Eru Cakra BDT (2016) Stock Price Prediction using Linear Regression based on Sentiment Analysis, International Conference on Advanced Computer Science and Information Systems, p. 147–154,
Fischer T, Krauss C. Deep learning with long short-term memory networks for financial market predictions. Euro J Operat Res. 2018;270:654–69.
Liu J et al. “Attention-Based Event Relevance Model for Stock Price Movement Prediction in China”, Conference on Knowledge Graph and Semantic Computing (2017)
Saud AS, Shakya S. Analysis of look back period for stock price prediction with RNN variants: A case study on banking sector of NEPSE. Procedia Comp Sci. 2020;167:788–98.
Shahi TB, Shrestha A, Neupane A, Guo W (2020) “Stock Price Forecasting with Deep Learning: A Comparative Study”, Multidisciplinary Digital Publishing Institute (MDPI)
Pai P-F, Lin C-S. A hybrid ARIMA and support vector machines model in stock price forecasting. Int J Manag Sci. 2005;33:497–505.
Cheong Fung GP, Xu Yu J, Lam W (2002) "News sensitive stock trends prediction," In: 6th Pacific-Asia Knowledge Discovery in Data Mining, Beijing
Lavrenko V , Schmill M, Lawrie D, Ogilvie P, Jensen D, Allan J (2000) "Mining of concurrent text and time series." In: Workshop of 6th International Conference on Knowledge Discovery and Data Mining
Li J, Bu H, Wu J. Sentiment-aware stock market prediction: a deep learning method. In Proceedings. Int Conf Ser Syst Ser Manag Dalian China. 2017;16:1–6.