arXiv q-fin· Jiahao Weng, Jiyu Liu·· 13 hr agoAI Score40
SHADE: A Pause-Alpha Stable Denoising-Based Shrinkage Method for Heavy-Tailed Probabilistic Time Series Forecasting
SHADE: Shrinkage via Halted Alpha-stable Denoising for Heavy-Tailed Probabilistic Time Series Forecasting
AI summary
The study proposes an Alpha-stable Lévy diffusion predictor called SHADE, which achieves distribution-level shrinkage through a residual stable kernel by reading outputs at non-zero depths, addressing the problem of overly narrow predictive distributions and overconfidence when Gaussian diffusion models forecast heavy-tailed time series.
Source:arXiv q-fin · arxiv.org