A comparison of some conformal quantile regression methods

Abstract

We compare two recently proposed methods that combine ideas from conformal inference and quantile regression to produce locally adaptive and marginally valid prediction intervals under sample exchangeability (Romano et al., 2019; Kivaranovic et al., 2019). First, we prove that these two approaches are asymptotically efficient in large samples, under some additional assumptions. Then we compare them empirically on simulated and real data. Our results demonstrate that the method in Romano et al. (2019) typically yields tighter prediction intervals in finite samples. Finally, we discuss how to tune these procedures by fixing the relative proportions of observations used for training and conformalization.

arXiv comment20 pages, 9 figures, 3 tables

References in corpus (3)

• Regression Conformal Prediction with Nearest Neighbours
• On-line predictive linear regression
• On-line predictive linear regression

Cited by in corpus (7)

• A review of predictive uncertainty estimation with machine learning
• Testing for Outliers with Conformal p-values
• Nested conformal prediction and quantile out-of-bag ensemble methods
• Valid prediction intervals for regression problems
• Improved conformalized quantile regression
• Uncertainty estimation of machine learning spatial precipitation predictions from satellite data
• Conformal inference for random objects