6 papers
Enforcing tail calibration when training probabilistic forecast models
Jakob Benjamin Wessel, Maybritt Schillinger, Frank Kwasniok +1
Probabilistic forecasts are typically obtained using state-of-the-art statistical and machine learning models, with model parameters estimated by optimizing a proper scoring rule o…
Residual Distribution Predictive Systems
Sam Allen, Enrico Pescara, Johanna Ziegel
Conformal predictive systems are sets of predictive distributions with theoretical out-of-sample calibration guarantees. The calibration guarantees are typically that the set of pr…
Assessing the conditional calibration of interval forecasts using decompositions of the interval score
Sam Allen, Julia Burnello, Johanna Ziegel
Forecasts for uncertain future events should be probabilistic. Probabilistic forecasts are commonly issued as prediction intervals, which provide a measure of uncertainty in the un…
Tail calibration of probabilistic forecasts
Sam Allen, Jonathan Koh, Johan Segers +1
Probabilistic forecasts comprehensively describe the uncertainty in the unknown future outcome, making them essential for decision making and risk management. While several methods…
In-sample calibration yields conformal calibration guarantees
Sam Allen, Georgios Gavrilopoulos, Alexander Henzi +2
Conformal predictive systems allow forecasters to issue predictive distributions for real-valued future outcomes that have out-of-sample calibration guarantees. On a more abstract…
Efficient pooling of predictions via kernel embeddings
Sam Allen, David Ginsbourger, Johanna Ziegel
Probabilistic predictions are probability distributions over the set of possible outcomes. Such predictions quantify the uncertainty in the outcome, making them essential for effec…