Trimmed Harrell-Davis quantile estimator based on the highest density interval of the given width

Traditional quantile estimators that are based on one or two order statistics a common way to estimate distribution quantiles the given samples. These robust, but their statistical efficiency is not always good enough. A more efficient alternative Harrell-Davis estimator which uses weighted sum of all statistics. Whereas this approach provides accurate estimations for light-tailed distributions, it's robust. To be able customize trade-off between and robustness, we could consider trimmed modification estimator. In approach, discard with low weights according highest density interval beta distribution.