tsbootstrap: Distribution-Free Uncertainty Quantification and Conformal Prediction for Time Series
arXiv:2607.06690v1 Announce Type: cross Abstract: Finance, sensing, and demand streams violate the exchangeability that IID conformal prediction and the IID bootstrap assume, and existing libraries implement either a general resampling engine or conformal calibration without the other. tsbootstrap provides block, residual, sieve, and wild resampling, classical bootstrap confidence intervals, and adaptive conformal calibrators (EnbPI, ACI, NexCP, AgACI) through a single typed API in which a specification object selects each method. In a controlled coverage study the IID bootstrap undercovers sh
The proliferation of time series data in domains like finance and sensing, coupled with the limitations of existing AI/ML tools, creates an immediate need for robust uncertainty quantification methods.
This development addresses a critical gap in AI/ML reliability for time-dependent data, providing more trustworthy predictions essential for high-stakes applications and reducing the risks associated with model deployment.
A new framework, tsbootstrap, enables more accurate and dependable forecasting and inference for time series, moving beyond the restrictive IID assumptions prevalent in current machine learning practices.
- · Quantitative finance
- · Predictive maintenance
- · Supply chain logistics
- · AI/ML developers
- · Legacy time series forecasting methods
- · Organizations relying solely on IID assumptions
Improved reliability and transparency of AI-driven predictions across various industries.
Increased adoption of conformal prediction and robust bootstrap methods in critical infrastructure and financial systems.
Potentially enables new regulatory frameworks around AI safety and interpretability for time-series applications.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI