Isotonic Distributional Regression
arXiv:1909.03725 · doi:10.1111/rssb.12450
Abstract
Isotonic distributional regression (IDR) is a powerful nonparametric technique for the estimation of conditional distributions under order restrictions. In a nutshell, IDR learns conditional distributions that are calibrated, and simultaneously optimal relative to comprehensive classes of relevant loss functions, subject to isotonicity constraints in terms of a partial order on the covariate space. Nonparametric isotonic quantile regression and nonparametric isotonic binary regression emerge as special cases. For prediction, we propose an interpolation method that generalizes extant specifications under the pool adjacent violators algorithm. We recommend the use of IDR as a generic benchmark technique in probabilistic forecast problems, as it does not involve any parameter tuning nor implementation choices, except for the selection of a partial order on the covariate space. The method can be combined with subsample aggregation, with the benefits of smoother regression functions and gains in computational efficiency. In a simulation study, we compare methods for distributional regression in terms of the continuous ranked probability score (CRPS) and estimation error, which are closely linked. In a case study on raw and postprocessed quantitative precipitation forecasts from a leading numerical weather prediction system, IDR is competitive with state of the art techniques.
References in corpus (11)
- OSQP: An Operator Splitting Solver for Quadratic Programs
- Neural networks for post-processing ensemble weather forecasts
- Quantile and Probability Curves Without Crossing
- Uncertainty Quantification in Complex Simulation Models Using Ensemble Copula Coupling
- Probabilistic Quantitative Precipitation Forecasting Using Ensemble Model Output Statistics
- Bayesian structured additive distributional regression with an application to regional income inequality in Germany
- Conditional Transformation Models
- Evaluating probabilistic classifiers: Reliability diagrams and score decompositions revisited
- Distributional Regression Forests for Probabilistic Precipitation Forecasting in Complex Terrain
- Forest-based methods and ensemble model output statistics for rainfall ensemble forecasting
- Monotone Least Squares and Isotonic Quantiles
Cited by in corpus (8)
- A review of predictive uncertainty estimation with machine learning
- Regression Diagnostics meets Forecast Evaluation: Conditional Calibration, Reliability Diagrams, and Coefficient of Determination
- Postprocessing of point predictions for probabilistic forecasting of day-ahead electricity prices: The benefits of using isotonic distributional regression
- Accelerating the pool-adjacent-violators algorithm for isotonic distributional regression
- Comparing Sequential Forecasters
- Distributional regression and its evaluation with the CRPS: Bounds and convergence of the minimax risk
- Estimation of a Likelihood Ratio Ordered Family of Distributions
- Post-processing of ensemble photovoltaic power forecasts with distributional and quantile regression methods