English

Approximation algorithms for confidence bands for time series

Machine Learning 2021-12-14 v1 Data Structures and Algorithms

Abstract

Confidence intervals are a standard technique for analyzing data. When applied to time series, confidence intervals are computed for each time point separately. Alternatively, we can compute confidence bands, where we are required to find the smallest area enveloping kk time series, where kk is a user parameter. Confidence bands can be then used to detect abnormal time series, not just individual observations within the time series. We will show that despite being an NP-hard problem it is possible to find optimal confidence band for some kk. We do this by considering a different problem: discovering regularized bands, where we minimize the envelope area minus the number of included time series weighted by a parameter α\alpha. Unlike normal confidence bands we can solve the problem exactly by using a minimum cut. By varying α\alpha we can obtain solutions for various kk. If we have a constraint kk for which we cannot find appropriate α\alpha, we demonstrate a simple algorithm that yields O(n)O(\sqrt{n}) approximation guarantee by connecting the problem to a minimum kk-union problem. This connection also implies that we cannot approximate the problem better than O(n1/4)O(n^{1/4}) under some (mild) assumptions. Finally, we consider a variant where instead of minimizing the area we minimize the maximum width. Here, we demonstrate a simple 2-approximation algorithm and show that we cannot achieve better approximation guarantee.

Keywords

Cite

@article{arxiv.2112.06225,
  title  = {Approximation algorithms for confidence bands for time series},
  author = {Nikolaj Tatti},
  journal= {arXiv preprint arXiv:2112.06225},
  year   = {2021}
}
R2 v1 2026-06-24T08:13:54.579Z