Assume we have a sample of size $n$ from an unspecified continuous distribution $F(\cdot)$. We wish to construct a tolerance interval to contain $(100\,\beta)\%$ of the population with a pre-specified confidence level $\gamma$ based on our sample. However, a conventional tolerance interval does not guarantee that the central $(100\,\beta)\%$ of the population is contained. In other words, a tolerance interval does not contain the $(1 - \beta)/2$ and $(1 + \beta)/2$ quantiles of the population distribution with confidence $\gamma$.
On the other hand: A tolerance interval that does contains the central $(100\,\beta)\%$ of the population with confidence $\gamma$ is called an equal-tailed$^{[1, 2]}$ or central tolerance interval. Meeker & Hahn (2017) call them "tolerance intervals to control both tails"$^{[3]}$ (section E5.2).
While Liu et al. (2021) detail the construction of a nonparametric tolerance interval, they explicitly omit the details of how to construct an equal-tailed nonparametric tolerance interval to save space in their paper.
Because tolerance intervals are synonymous with confidence intervals for percentiles, would one possibility be to calculate a one-sided lower and upper confidence interval for the $0.025$ and $0.975$ percentiles, respectively? I also found the paper by Hayter (2014)$^{[4]}$ that describes a method to calculate simultaneous nonparametric confidence intervals for percentiles but I was not able to implement the proposed algorithm in R
and test it.
Question: How can the calculations shown here be modified so that the resulting nonparametric tolerance interval contains the central $(100\,\beta)\%$ of the unknown continuous population distribution with confidence $\gamma$?
References
$[1]$: Liu, W., Bretz, F., & Cortina-Borja, M. (2021). Reference range: Which statistical intervals to use?. Statistical methods in medical research, 30(2), 523-534. (link)
$[2]$: Jan, S. L., & Shieh, G. (2018). The Bland-Altman range of agreement: Exact interval procedure and sample size determination. Computers in biology and medicine, 100, 247-252. (link)
$[3]$: Meeker, W. Q., Hahn, G. J., & Escobar, L. A. (2017). Statistical intervals: a guide for practitioners and researchers. 2nd ed. John Wiley & Sons. (link)
$[4]$: Hayter, A. J. (2014). Simultaneous confidence intervals for several quantiles of an unknown distribution. The American Statistician, 68(1), 56-62. (link)