OrdMonReg: Compute least squares estimates of one bounded or two ordered isotonic regression curves

We consider the problem of estimating two isotonic regression curves g1* and g2* under the constraint that they are ordered, i.e. g1* <= g2*. Given two sets of n data points y_1, ..., y_n and z_1, ..., z_n that are observed at (the same) deterministic design points x_1, ..., x_n, the estimates are obtained by minimizing the Least Squares criterion L(a, b) = sum_{i=1}^n (y_i - a_i)^2 w1(x_i) + sum_{i=1}^n (z_i - b_i)^2 w2(x_i) over the class of pairs of vectors (a, b) such that a and b are isotonic and a_i <= b_i for all i = 1, ..., n. We offer two different approaches to compute the estimates: a projected subgradient algorithm where the projection is calculated using a PAVA as well as Dykstra's cyclical projection algorithm.

Version: 1.0.3
Published: 2011-12-01
Author: Fadoua Balabdaoui, Kaspar Rufibach, Filippo Santambrogio
Maintainer: Kaspar Rufibach <kaspar.rufibach at gmail.com>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
URL: http://www.ceremade.dauphine.fr/~fadoua, http://www.kasparrufibach.ch, http://www.math.u-psud.fr/~santambr/
NeedsCompilation: no
Materials: NEWS
CRAN checks: OrdMonReg results

Downloads:

Reference manual: OrdMonReg.pdf
Package source: OrdMonReg_1.0.3.tar.gz
Windows binaries: r-devel: OrdMonReg_1.0.3.zip, r-release: OrdMonReg_1.0.3.zip, r-oldrel: OrdMonReg_1.0.3.zip
OS X El Capitan binaries: r-release: OrdMonReg_1.0.3.tgz
OS X Mavericks binaries: r-oldrel: OrdMonReg_1.0.3.tgz
Old sources: OrdMonReg archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=OrdMonReg to link to this page.