Saddlepoint Confidence Intervals For Directly Standardized Rates
Abstract
We develop a novel method of confidence interval construction for directly standardized rates. These intervals involve saddlepoint approximations to the intractable distribution of a weighted sum of Poisson random variables and the determination of hypothetical Poisson mean values for each of the age groups. Simulation studies show that, in terms of coverage probability and length, the saddlepoint confidence interval outperforms four competing confidence intervals obtained from the moment matching, gamma-based and ABC bootstrap methods. Robustness simulation studies show that our proposed procedure essentially performed no worse than the competing procedures in terms of coverage probability or length. Finally, we consider a real-world application involving data from the WHO MONICA Project.
Recommended Citation
P. M. Edirisinghe and R. L. Paige, "Saddlepoint Confidence Intervals For Directly Standardized Rates," Communications in Statistics Simulation and Computation, Taylor and Francis Group; Taylor and Francis, Jan 2025.
The definitive version is available at https://doi.org/10.1080/03610918.2025.2519502
Department(s)
Mathematics and Statistics
Keywords and Phrases
Causal inference; Confidence intervals; Directly standardized rates; Saddlepoint approximations
International Standard Serial Number (ISSN)
1532-4141; 0361-0918
Document Type
Article - Journal
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2025 Taylor and Francis Group; Taylor and Francis, All rights reserved.
Publication Date
01 Jan 2025
