Kernel-based Optimally Weighted Conformal Time-Series Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lee, Jonghyeok, Xu, Chen, Xie, Yao
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912897306722304
author Lee, Jonghyeok
Xu, Chen
Xie, Yao
author_facet Lee, Jonghyeok
Xu, Chen
Xie, Yao
contents In this work, we present a novel conformal prediction method for time-series, which we call Kernel-based Optimally Weighted Conformal Prediction Intervals (KOWCPI). Specifically, KOWCPI adapts the classic Reweighted Nadaraya-Watson (RNW) estimator for quantile regression on dependent data and learns optimal data-adaptive weights. Theoretically, we tackle the challenge of establishing a conditional coverage guarantee for non-exchangeable data under strong mixing conditions on the non-conformity scores. We demonstrate the superior performance of KOWCPI on real and synthetic time-series data against state-of-the-art methods, where KOWCPI achieves narrower confidence intervals without losing coverage.
format Preprint
id arxiv_https___arxiv_org_abs_2405_16828
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Kernel-based Optimally Weighted Conformal Time-Series Prediction
Lee, Jonghyeok
Xu, Chen
Xie, Yao
Machine Learning
Statistics Theory
In this work, we present a novel conformal prediction method for time-series, which we call Kernel-based Optimally Weighted Conformal Prediction Intervals (KOWCPI). Specifically, KOWCPI adapts the classic Reweighted Nadaraya-Watson (RNW) estimator for quantile regression on dependent data and learns optimal data-adaptive weights. Theoretically, we tackle the challenge of establishing a conditional coverage guarantee for non-exchangeable data under strong mixing conditions on the non-conformity scores. We demonstrate the superior performance of KOWCPI on real and synthetic time-series data against state-of-the-art methods, where KOWCPI achieves narrower confidence intervals without losing coverage.
title Kernel-based Optimally Weighted Conformal Time-Series Prediction
topic Machine Learning
Statistics Theory
url https://arxiv.org/abs/2405.16828