Stock Price Prediction Based on TimesNet

Data Collection

The stock data is collected using the open-source financial interface AkShare, retrieving relevant stock market data. The dataset used in this project originates from East Money (东方财富网) - Market Homepage - Shanghai, Shenzhen, and Beijing A-Shares - Daily Market Data.

Data Preprocessing

Data Cleaning: Handling missing values, outliers, and duplicate data.

Data Transformation: Standardization, normalization, and encoding categorical variables.

Feature Engineering: Creating new features, selecting important features, and performing dimensionality reduction.

Model Training and Evaluation

The model is trained using the training dataset, while validation and test datasets are used to assess performance. The evaluation metrics used include:

Mean Absolute Error (MAE)、Mean Squared Error (MSE)、Root Mean Squared Error (RMSE)、Mean Absolute Percentage Error (MAPE)、Mean Squared Percentage Error (MSPE)

This project leverages TimesNet（https://github.com/thuml/Time-Series-Library） for time-series stock price prediction, focusing on accurate market trend forecasting. If you want to know more about the model, please refer to the Timesent official documentation.