Research on stock price prediction based on NeuralForecast: A case study of CSI 300 index components
Abstract
LSTM and NHITS, we predict the prices of 182 CSI 300 stocks over the next 48 trading days, based on data from 2019 to 2023. Data processing involves logarithms, first-order differencing, and moving averages. NHITS outperforms LSTM, especially with moving averages
and logarithmic price features. The results suggest NHITS provides more accurate and robust predictions, highlighting the potential of deep
learning in stock forecasting.
Keywords
Full Text:
PDFReferences
[1] Risheng Qiao. (2023). Research on portfolio optimization based on deep learning quantitative analysis (Doctoral dissertation,
Tianjin University of Technology). doi:10.27360/d.cnki.gtlgy.2023.001274.
[2] Xufan Bao. (2023). Research on quantitative investment strategy of funds based on ensemble learning (Master’s thesis, Soochow
University). doi:10.27351/d.cnki.gszhu.2023.001227.
[3] Zixian Yang. (2022). Research on quantitative stock selection and effect evaluation based on LSTM algorithm (Master’s thesis,
Shenzhen University). doi:10.27321/d.cnki.gszdu.2022.001877.
[4] Hailong Liu, Lihui Zheng, & Chongfeng Wu. (2001). Progress review of modern financial theory. Systems Engineering - Theory
& Practice, 21(1), 14-20+40.
[5] Xiangyi Chen. (2018). Prediction of CSI 300 index based on convolutional neural network (Master’s thesis, Beijing University of
Posts and Telecommunications). doi:10.27461/d.cnki.gzjdx.2019.000444.
[6] Jiawei Mo. (2023). Research on stock price prediction and stock selection strategy based on LSTM model (Master’s thesis, Guangzhou University). doi:10.27040/d.cnki.ggzdu.2023.001786.
Volume 9 Issue 6 -55-
[7] Zhihui Huang. (2019). Research on quantitative stock selection model based on convolutional neural network (Master’s thesis,
Zhejiang University). doi:10.27461/d.cnki.gzjdx.2019.000444.
[8] Yu Wei, Xiaodong Lai, & Jiang Yu. (2013). Research on the intra-day hedging model and efficiency of CSI 300 stock index futures. Journal of Management Science, 16(3), 29-40.
[9] Markowitz M H.Portfolio selection,efficient diversification of investment[M].Yale UniversityPress,195Sharpe W.Capital asset
prices:a theory of market equilibrium under conditions of risk[J].Journal of Finance,1964,19:425~ 442 .
[10]Lintner J.The valuation of risk assets and the selection of risky investments in stock portfolios and capital budgets.Review of Economics and Statistics,1 965,47(2 ) :1 3~ 37.
[11]Mossin,J.Equilibrium in a capital asset market[J].Econometrica,1 966,34(4) :768~ 783.
[12]Roberts H.Statistical versus clinical prediction of the stock market.Unpublished Manuscript[D],CRSP,Chicago:University of Chicago,May,1967.
[13]Fama E.The behavior of stock market prices[J].Journal of Business,1 965,38(1 ) :34~ 1 0 5.
[14]Li, T., Liu, Z., Shen, Y., Wang, X., Chen, H., & Huang, S. (2023). MASTER: Market-Guided Stock Transformer for Stock Price
Forecasting. AAAI Conference on ArtificiSal Intelligence.
[15]Qian, H., Zhou, H., Zhao, Q., Chen, H., Yao, H., Wang, J., Liu, Z., Yu, F., Zhang, Z., & Zhou, J. (2024). MDGNN: Multi-Relational Dynamic Graph Neural Network for Comprehensive and Dynamic Stock Investment Prediction. ArXiv, abs/2402.06633.
DOI: http://dx.doi.org/10.18686/fm.v9i6.13664
Refbacks
- There are currently no refbacks.
