Abstract
The small sample properties of a number of approaches to finding a confidence interval for a slope parameter in rank regression are investigated. An approach based on the bootstrap percentile-t procedure is shown to have excellent overall performance.
Original language | English |
---|---|
Pages (from-to) | 13-22 |
Number of pages | 10 |
Journal | Journal of Statistical Computation and Simulation |
Volume | 53 |
Issue number | 1-2 |
DOIs | |
State | Published - Oct 1 1995 |
Keywords
- Bootstrap
- Jackknife
- Monte Carlo
- Standard error
ASJC Scopus subject areas
- Statistics and Probability
- Modeling and Simulation
- Statistics, Probability and Uncertainty
- Applied Mathematics