Research Article
Research on Stock Price Prediction Based on Hidden Markov Model and Elastic Feedback Algorithm
@INPROCEEDINGS{10.4108/eai.8-9-2023.2340080, author={Na Jing and Shuang Li and Longdong Wang}, title={Research on Stock Price Prediction Based on Hidden Markov Model and Elastic Feedback Algorithm}, proceedings={Proceedings of the 4th International Conference on Modern Education and Information Management, ICMEIM 2023, September 8--10, 2023, Wuhan, China}, publisher={EAI}, proceedings_a={ICMEIM}, year={2023}, month={11}, keywords={stocks forecast; hidden markov model; elastic feedback algorithm}, doi={10.4108/eai.8-9-2023.2340080} }- Na Jing
Shuang Li
Longdong Wang
Year: 2023
Research on Stock Price Prediction Based on Hidden Markov Model and Elastic Feedback Algorithm
ICMEIM
EAI
DOI: 10.4108/eai.8-9-2023.2340080
Abstract
In the current financial market, predicting stock prices accurately is a major challenge. To address this, we propose a new hybrid algorithm. The stock market is known for its chaotic nature, and the hidden Markov model is a good fit for its current characteristics. However, due to the vast amount of market data and its randomness, a single Markov method isn't sufficient for accurate price forecasting. In our approach, we enhance the forecasting method with an elastic feedback algorithm. We first classify forecasts into three states: rising, falling, and staying relatively stable, using the hidden Markov model. Then, we backtest these predictions against the actual prices from the last 20 days. By incorporating the elastic feedback algorithm, we significantly improve forecasting accuracy from 60%, 65%, and 55% to 75%, 75%, 60%.
