Academic Journal of Business & Management, 2022, 4(3); doi: 10.25236/AJBM.2022.040308.
Ziheng Zhang
School of Ande, Xi' an University of Architecture and Technology, Xi’an, 710311, China
When technical methods are used to establish LSTM stock prediction model, traditional methods often lead to poor generalization and poor prediction effect due to many input data variables selected, data information overlap, outliers have a great influence on training and other factors. To solve these problems, this paper proposes to use principle component analysis method to reduce the dimension of basic data, combine with stock related technical indicators KDJ and MACD as input data, and make prediction after adjusting the model according to stock characteristics. The experimental results show that the PCA-LSTM model reduces the average error of prediction, greatly reduces the running time, improves the stability of prediction, and more accurately predicts the closing price of Yaxing, which has application value.
LSTM model, Stock price prediction, Neural network, Principal component analysis
Ziheng Zhang. Research on Stock Price Prediction Based on PCA-LSTM Model. Academic Journal of Business & Management (2022) Vol. 4, Issue 3: 42-47. https://doi.org/10.25236/AJBM.2022.040308.
[1] Yang Q, Cao Xianbing. Stock price analysis and forecasting based on ARMA-GARCH model [J]. Practice and understanding of mathematics, 2016, 46(06): 80-86.
[2] Huang LJ, Jin TX. A study of Islamic stock market stability based on EGARCH-M model [J]. Gansu Journal of Theory, 2019, (06): 107-115+2.
[3] Fang J. An empirical study of VaR measures in Chinese stock market: A semi-parametric approach based on IGARCH [J]. Financial Theory and Teaching, 2018, (03): 15-18.
[4] Jing Nan, Wu Zhao, Wang Hefei. A hybrid model integrating deep learning with investor sentiment analysis for stock price prediction [J]. Expert Systems with Applications, 2021, 178.
[5] Cao Xiao, Sun Hongbing. Stock price prediction based on gray GARCH model and BP neural network [J]. Software, 2017, 38(11): 126-131.
[6] Deng Xuan-Kun, Wan-Liang, Huang Na-Na. Research on stock prediction based on DAE-BP neural network [J]. Computer Engineering and Applications, 2019, 55(03): 126-132.
[7] Budiharto Widodo. Data science approach to stock prices forecasting in Indonesia during Covid-19 using Long Short-Term Memory (LSTM) [J]. Journal of Big Data, 2021, 8(1).
[8] Ko ChingRu, Chang HsienTsung. LSTM-based sentiment analysis for stock price forecast. [J]. PeerJ. Computer science, 2021, 7.
[9] Applied Intelligence; Researchers from University of Science and Technology Beijing Detail New Studies and Findings in the Area of Applied Intelligence (A Hybrid Stock Price Index Forecasting Model Based on Variational Mode Decomposition and Lstm Network) [J]. Journal of Robotics& Machine Learning, 2020.