bootUR: An R Package for Bootstrap Unit Root Tests

Abstract

Unit root tests form an essential part of any time series analysis. We provide practitioners with a single, unified framework for comprehensive and reliable unit root testing in the R package bootUR. The package’s backbone is the popular augmented DickeyFuller test paired with a union of rejections principle, which can be performed directly on single time series or multiple (including panel) time series. Accurate inference is ensured through the use of bootstrap methods. The package addresses the needs of both novice users, by providing user-friendly and easy-to-implement functions with sensible default options, as well as expert users, by giving full user-control to adjust the tests to one’s desired settings. Our parallelized C++ implementation ensures that all unit root tests are scalable to datasets containing many time series.