A Fractal Feature-Based Model for Predicting Financial Price Trends

The complexity and unpredictability of financial markets have been the focus of attention for investors and researchers. Traditional financial time series analysis methods have limitations in dealing with nonlinear and non-stationary data, while fractal theory, as a tool for studying complex systems, shows potential advantages in financial price trend forecasting. This study aims to investigate the effectiveness and application of fractal features in financial price trend forecasting. Firstly, the study reviews the basic concepts of fractal theory and resamples financial data based on fractal features, secondly, the study proposes a financial price trend prediction model incorporating fractal features and predicts future price trends by slipping the specified categorical voting rules. In addition, the study explores how to combine fractal features with traditional technical analysis tools and machine learning algorithms to improve the accuracy and robustness of the forecasts. Finally, empirical analyses are conducted with real data from financial markets, and the results show that fractal features can effectively reveal the non-linear patterns of price fluctuations and can effectively reveal the self-similarity characteristics of financial markets. We expect to provide a new perspective and tool for financial market analysis and investment decision-making to cope with the uncertainty and complexity of financial markets.