Prediction of Corporate Bond Prices Based on Machine Learning Algorithms
Download as PDF
DOI: 10.23977/ICAMCS2022.006
Corresponding Author
Chiba Naoto
ABSTRACT
This article uses whether several state-of-the-art machine learning or deep learning methods (GBDT models, NN, etc.) can be well applied to financial applications. We conduct comprehensive experiments on corporate bond price forecasting using several different computational models and provide an in-depth analysis and comparison of their performance. Furthermore, our results suggest that deep learning methods may not always be omnipotent when the data space is rather low/sparse, which is consistent with our general intuition and previous literature.
KEYWORDS
Machine learning, Deep learning, Bonds