Extension of the pool-adjacent-violators algorithm
Dei‐In Tang, Shang P. Lin · Communication in Statistics- Theory and Methods · 1991
The pool-adjacent-violators algorithm (PAVA) is an efficient algorithm which converges in a finite number of steps. However, it has been applicable so far only in isotonic regression with the simple order. This report extends its applicability to other quadratic programming problems, including certain one-sided multivariate testing problems and concave regression problems.