Author(s)

Jian Zhang

Corresponding Author

Jian Zhang

Abstract

Stock prediction has always been a hot and difficult topic for scholars at home and abroad, and stock research also has important practical and theoretical significance. Stock data is sequential. However, neural networks have performed well in dealing with time series problems. Among them, long short-term memory neural networks are very suitable for processing this time series data with long-term dependence, so this article is mainly based on principal component analysis and performs feature extraction on stock data. The stock trend is analyzed and predicted through the adjustment and optimization of the LSTM neural network model.

Keywords

Stock price prediction; LSTM; Principal component analysis; Neural network