The paper introduces a generative probabilistic forecasting method, WIAE-GPF, leveraging weak innovation representation for accurate predictions in dynamic market scenarios. The study compares various forecasting techniques and highlights the effectiveness of nonparametric approaches like WIAE-GPF over parametric models. Results show that WIAE-GPF outperforms other methods in both point and probabilistic forecasting metrics across different market applications. The simplicity and Bayesian sufficiency of the weak innovation representation contribute to the success of WIAE-GPF in capturing volatile market dynamics.
To Another Language
from source content
arxiv.org
Key Insights Distilled From
by Xinyi Wang,L... at arxiv.org 03-12-2024
https://arxiv.org/pdf/2403.05743.pdfDeeper Inquiries