Abstract
The isotonic median regression problem arises from statistics. An algorithm, the PAV algorithm, has been proposed for solving this problem since 1980. In this paper, we propose two kinds of data structures for efficiently implementing the PAV algorithm. The running time of the algorithm is also improved.
Original language | English (US) |
---|---|
Pages (from-to) | 67-70 |
Number of pages | 4 |
Journal | Applied Mathematics Letters |
Volume | 8 |
Issue number | 2 |
DOIs | |
State | Published - Mar 1995 |
Externally published | Yes |
Keywords
- Balanced search tree
- Median
- Merging
- Running time
ASJC Scopus subject areas
- Applied Mathematics