Abstract
We propose a new algorithm for envelope estimation, along with a new n-consistent method for computing starting values. The new algorithm, which does not require optimization over a Grassmannian, is shown by simulation to be much faster and typically more accurate than the best existing algorithm proposed by Cook and Zhang (2016).
Original language | English (US) |
---|---|
Pages (from-to) | 42-54 |
Number of pages | 13 |
Journal | Journal of Multivariate Analysis |
Volume | 150 |
DOIs | |
State | Published - Jan 1 2016 |
Fingerprint
Keywords
- Envelopes
- Grassmann manifold
- Reducing subspaces
Cite this
A note on fast envelope estimation. / Cook, R. Dennis; Forzani, Liliana; Su, Zhihua.
In: Journal of Multivariate Analysis, Vol. 150, 01.01.2016, p. 42-54.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A note on fast envelope estimation
AU - Cook, R. Dennis
AU - Forzani, Liliana
AU - Su, Zhihua
PY - 2016/1/1
Y1 - 2016/1/1
N2 - We propose a new algorithm for envelope estimation, along with a new n-consistent method for computing starting values. The new algorithm, which does not require optimization over a Grassmannian, is shown by simulation to be much faster and typically more accurate than the best existing algorithm proposed by Cook and Zhang (2016).
AB - We propose a new algorithm for envelope estimation, along with a new n-consistent method for computing starting values. The new algorithm, which does not require optimization over a Grassmannian, is shown by simulation to be much faster and typically more accurate than the best existing algorithm proposed by Cook and Zhang (2016).
KW - Envelopes
KW - Grassmann manifold
KW - Reducing subspaces
UR - http://www.scopus.com/inward/record.url?scp=84979762838&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84979762838&partnerID=8YFLogxK
U2 - 10.1016/j.jmva.2016.05.006
DO - 10.1016/j.jmva.2016.05.006
M3 - Article
AN - SCOPUS:84979762838
VL - 150
SP - 42
EP - 54
JO - Journal of Multivariate Analysis
JF - Journal of Multivariate Analysis
SN - 0047-259X
ER -