# Selective Time Series Forecasting via Metalearning

arXiv: https://arxiv.org/abs/2606.23448 · HTML: https://mlsanity.com/paper/2606.23448/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 28 references that are in the map.
- First posted on arXiv 22 Jun 2026.

## Builds on (most cited first, 10 of 10)

- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [KAN: Kolmogorov-Arnold Networks](https://mlsanity.com/paper/2404.19756/) (2024)
- [GIFT-Eval: A Benchmark For General Time Series Forecasting Model Evaluation](https://mlsanity.com/paper/2410.10393/) (2024)
- [Forecast Evaluation for Data Scientists: Common Pitfalls and Best Practices](https://mlsanity.com/paper/2203.10716/) (2022)
- [Criteria for Classifying Forecasting Methods](https://mlsanity.com/paper/2212.03523/) (2022)
- [Selective Regression Under Fairness Criteria](https://mlsanity.com/paper/2110.15403/) (2021)
- [Ensembles of Localised Models for Time Series Forecasting](https://mlsanity.com/paper/2012.15059/) (2020)
- [VEST: Automatic Feature Engineering for Forecasting](https://mlsanity.com/paper/2010.07137/) (2020)
- [Classification with Reject Option: Distribution-free Error Guarantees via Conformal Prediction](https://mlsanity.com/paper/2506.21802/) (2025)
- [Towards Reliable Time Series Forecasting under Future Uncertainty: Ambiguity and Novelty Rejection Mechanisms](https://mlsanity.com/paper/2503.19656/) (2025)

## Built on by (most cited first, 1 of 1)

- [When Should a World Model Move? Loss-Conditioned State Execution](https://mlsanity.com/paper/2609.15801/) (2026)

## Abstract (arXiv)

Deep learning methods have achieved state-of-the-art in time series forecasting, yet their accuracy varies considerably across samples, as some instances remain inherently difficult to predict. Reject option mechanisms, which allow models to abstain from high-risk predictions, are well established in classification and regression but underexplored in forecasting. Existing abstention strategies typically rely on proxies, such as the width of the prediction interval or learned confidence scores derived from forecasts. However, these approaches are inherently tied to the training domain, limiting their ability to generalize. We propose a selective forecasting framework that addresses this limitation by modeling the empirical percentile of forecasting errors, that is, a scale-invariant statistic, based on structural characteristics extracted from recent lags via metalearning. By decoupling the rejection decision from the forecast itself and grounding it in domain-agnostic features, the framework enables effective abstention transfer across heterogeneous time series. Experiments in both in-domain and transfer learning settings show that rejecting samples predicted as challenging consistently improves forecasting accuracy across coverage levels.