Extension of the pool-adjacent-violators algorithm |
| |
Authors: | Dei- In Tang Shang P. Lin |
| |
Affiliation: | Nathan S. Kline Institute for Psychiatric Research , New York, Orangeburg, 10962, U.S.A |
| |
Abstract: | 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. |
| |
Keywords: | concave regression Fenchel duality isotonic regression one-sided multivariate tests qundratic programming |
|
|