Issue |
ITM Web Conf.
Volume 70, 2025
2024 2nd International Conference on Data Science, Advanced Algorithm and Intelligent Computing (DAI 2024)
|
|
---|---|---|
Article Number | 04017 | |
Number of page(s) | 9 | |
Section | AI and Advanced Applications | |
DOI | https://doi.org/10.1051/itmconf/20257004017 | |
Published online | 23 January 2025 |
Forecasting the Treasury Yield Spread for FRED T10Y2Y Data Based on Multiple Approaches
American International School of Cape Town, Cape Town, South Africa
Corresponding author: yleu@aisct.org
As a matter of fact, price prediction for financial underlying assets always been a hot topic in financial fields in recent years with high volatility. With this in mind. this study looks into the usage of machine learning models to predict the yield spread between 10-year and 2-year US Treasury bonds (T10Y2Y). Based on the data from the Federal Reserve Economic Data (FRED) database (1976-2024), this study assesses the performance of four different models: multi-layer perceptron (MLP) regression, LSTM, ARIMA model and Facebook Prophet model. Each model’s performance is measured using MAE, MSE, RMSE as well as F2 score. The results show that both MLP regression and LSTM models achieve high accuracy in predicting the yield spread. However, MLP regression outperforms LSTM in terms of producing more reasonable future predictions, particularly over longer time periods. ARIMA and Prophet, while effective for linear forecasts, were confused by the data and made unreasonable and incorrect predictions
© The Authors, published by EDP Sciences, 2025
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.