ASE: A modular framework to integrate text in time series forecasting
In brief
- ASE merges textual and numerical data to improve time series forecasting accuracy.
- On TESS datasets, ASE reduces MSE compared to published scores.
- On CSI 300, ASE boosts RankIC by 0.000642 over the numerical-only model.
The ASE framework enhances time series forecasting by incorporating external textual data. It evaluates their conditional contribution and optimizes their combination with numerical models. Tests on four TESS datasets and a stock ranking show significant improvements over existing methods.
What it changes for you
- Finance & BankingStock forecasting models could integrate textual data to refine predictions and reduce errors.
This summary gives you the essentials. The full article, its figures and quotes are on cs.AI updates on arXiv.org.
Read the original articleAI-generated summary of the source, checked by rules. Report an error