Iterative Reweighted Partial Least Squares Estimation for GLMs
1995
We extend the concept of partial least squares (PLS) into the framework of generalized linear models. These models form a sequence of rank one approximations useful for predicting the response variable when the explanatory information is severely ill-conditioned or ill-posed. An Iter ative reweighted PLS algorithm is presented with various theoretical prop erties. Connections to principal component and maximum likelihood esti mation are made, as well as suggestions for rules to choose the proper rank of the final model.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
7
References
0
Citations
NaN
KQI