We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Sequential hypothesis testing for selecting the number of changepoints in segmented regression models.
- Authors
Priulla, Andrea; D'Angelo, Nicoletta
- Abstract
Segmented regression is widely used in many disciplines, especially when dealing with environmental data. This paper deals with the problem of selecting the correct number of changepoints in segmented regression models. A review of the usual selection criteria, namely information criteria and hypothesis testing, is provided. We enhance the latter method by proposing a novel sequential hypothesis testing procedure to address this problem. Our sequential procedure's performance is compared to methods based on information-based criteria through simulation studies. The results show that our proposal performs similarly to its competitors for the Gaussian, Binomial, and Poisson cases. Finally, we present two applications to environmental datasets of crime data in Valencia and global temperature land data.
- Subjects
VALENCIA (Spain); REGRESSION analysis; LAND surface temperature; ENVIRONMENTAL crimes; HYPOTHESIS
- Publication
Environmental & Ecological Statistics, 2024, Vol 31, Issue 2, p583
- ISSN
1352-8505
- Publication type
Article
- DOI
10.1007/s10651-024-00605-x