arrow
Volume 24, Issue 6
Data Preordering in Generalized PAV Algorithm for Monotonic Regression

Oleg Burdakov, Anders Grimvall & Oleg Sysoev

J. Comp. Math., 24 (2006), pp. 771-790.

Published online: 2006-12

Export citation
  • Abstract

Monotonic regression (MR) is a least distance problem with monotonicity constraints induced by a partially ordered data set of observations. In our recent publication [In Ser. Nonconvex Optimization and Its Applications, Springer-Verlag, (2006) 83, pp. 25-33], the Pool-Adjacent-Violators algorithm (PAV) was generalized from completely to partially ordered data sets (posets). The new algorithm, called GPAV, is characterized by the very low computational complexity, which is of second order in the number of observations. It treats the observations in a consecutive order, and it can follow any arbitrarily chosen topological order of the poset of observations. The GPAV algorithm produces a sufficiently accurate solution to the MR problem, but the accuracy depends on the chosen topological order. Here we prove that there exists a topological order for which the resulted GPAV solution is optimal. Furthermore, we present results of extensive numerical experiments, from which we draw conclusions about the most and the least preferable topological orders.

  • AMS Subject Headings

  • Copyright

COPYRIGHT: © Global Science Press

  • Email address
  • BibTex
  • RIS
  • TXT
@Article{JCM-24-771, author = {Oleg Burdakov, Anders Grimvall and Oleg Sysoev}, title = {Data Preordering in Generalized PAV Algorithm for Monotonic Regression}, journal = {Journal of Computational Mathematics}, year = {2006}, volume = {24}, number = {6}, pages = {771--790}, abstract = {

Monotonic regression (MR) is a least distance problem with monotonicity constraints induced by a partially ordered data set of observations. In our recent publication [In Ser. Nonconvex Optimization and Its Applications, Springer-Verlag, (2006) 83, pp. 25-33], the Pool-Adjacent-Violators algorithm (PAV) was generalized from completely to partially ordered data sets (posets). The new algorithm, called GPAV, is characterized by the very low computational complexity, which is of second order in the number of observations. It treats the observations in a consecutive order, and it can follow any arbitrarily chosen topological order of the poset of observations. The GPAV algorithm produces a sufficiently accurate solution to the MR problem, but the accuracy depends on the chosen topological order. Here we prove that there exists a topological order for which the resulted GPAV solution is optimal. Furthermore, we present results of extensive numerical experiments, from which we draw conclusions about the most and the least preferable topological orders.

}, issn = {1991-7139}, doi = {https://doi.org/}, url = {http://global-sci.org/intro/article_detail/jcm/8790.html} }
TY - JOUR T1 - Data Preordering in Generalized PAV Algorithm for Monotonic Regression AU - Oleg Burdakov, Anders Grimvall & Oleg Sysoev JO - Journal of Computational Mathematics VL - 6 SP - 771 EP - 790 PY - 2006 DA - 2006/12 SN - 24 DO - http://doi.org/ UR - https://global-sci.org/intro/article_detail/jcm/8790.html KW - Quadratic programming, Large scale optimization, Least distance problem, Monotonic regression, Partially ordered data set, Pool-adjacent-violators algorithm. AB -

Monotonic regression (MR) is a least distance problem with monotonicity constraints induced by a partially ordered data set of observations. In our recent publication [In Ser. Nonconvex Optimization and Its Applications, Springer-Verlag, (2006) 83, pp. 25-33], the Pool-Adjacent-Violators algorithm (PAV) was generalized from completely to partially ordered data sets (posets). The new algorithm, called GPAV, is characterized by the very low computational complexity, which is of second order in the number of observations. It treats the observations in a consecutive order, and it can follow any arbitrarily chosen topological order of the poset of observations. The GPAV algorithm produces a sufficiently accurate solution to the MR problem, but the accuracy depends on the chosen topological order. Here we prove that there exists a topological order for which the resulted GPAV solution is optimal. Furthermore, we present results of extensive numerical experiments, from which we draw conclusions about the most and the least preferable topological orders.

Oleg Burdakov, Anders Grimvall and Oleg Sysoev. (2006). Data Preordering in Generalized PAV Algorithm for Monotonic Regression. Journal of Computational Mathematics. 24 (6). 771-790. doi:
Copy to clipboard
The citation has been copied to your clipboard