GDCI: Robust empirical Bayes confidence intervals using gamma-divergence

This repository provides R code implementing robust empirical Bayes confidence intervals using gamma-divergence, as proposed by the following paper.

Kurisu, D., Ishihara, T. and Sugasawa, S. (2021). Adaptively Robust small area estimation: balancing robustness and efficiency of empirical Bayes confidence intervals. arXiv:2108.11551

The repository includes the following files.

• GDCI-function.R : Script implementing the proposed method
• simulation.R : Script of one-shot simulation
• crimedata.RData : Summary statistics of crimedata used in our analysis
• data-analysis.R : Script of an application to the crime data